So far we've supported OpenAI and we will soon start the basic income project. We have a few ideas about what to do next, but we'd love to hear from the community about what we should be considering. Thanks!
Bacteriophages. Antibiotics are running out, while phages have been tested by the Soviets, and are available as an over-the-counter treatment in certain countries.
Feynman’s and Drexler’s molecular nanotechnology. Smalley muddied the waters, and the National Research Council dropped the ball. We can do better than just nanoelectronics.
See also “Ask PG: What Is The Most Frighteningly Ambitious Idea You Have Been Pitched On?”, paying attention to the Yudkowsky Ambition scale: https://news.ycombinator.com/item?id=4509934
I couldn't agree more about funding SENS related research.
I think focusing on extended lifespan misses the point though (although it's a nice side benefit).
The diseases of old age cause more suffering than everything else combined.
Seeing people I love deteriorate to a state of constant suffering is unbearable to watch. Seeing signs of ageing in myself hammers home just how helpless I am to do anything.
I don't understand why AI safety gets more attention. We're already facing our own personal dissolution from natural causes -- a real and tangible existential threat which is already claiming people in the most torturous ways possible. To be brutally honest, I'd rather die quickly in an AI takeover than suffer decades of chronic pain and dementia... and one of these scenarios seems a lot more likely than the other.
Let's get real, take responsibility, and do something about ageing.
I don't think this point gets mentioned enough: since physical degeneration due to aging is such a big risk factor for so many diseases, anything we can do to reverse that degeneration could end up being hugely effective at preventing some of the diseases of old age. And there's a surprising lack of investment in this, probably because discussions tend to get bogged down in lengthy, heated philosophical digressions about how long people should want to live.
If you're looking for medical miracles, this is a very plausible place to find them.
Reducing existential risk is not about minimizing the amount of suffering in our own lives. It's about making the lives of our potential trillions of descendants possible.
There's no reason to care about "potential trillions of descendants" who don't exist, nor is there any reason to maximize the amount of humans in the universe, etc.
The "existential risk" talk tends to assume, to start with, that some wonderful, incredibly happy and nice "intergalactic civilization" of future transhumans or posthumans will exist, unless existential risks "destroy" it "in embryo" in our near-term future. Given the incredible prior improbability of intergalactic utopias, there's basically no reason to take the thought-experiment at its word or to value a particularly improbable imagined far-future over the actually-existing present and very likely near-futures.
The "existential risk" talk tends to assume, to start with, that some wonderful, incredibly happy and nice "intergalactic civilization" of future transhumans or posthumans will exist...
Can you cite such a person? I've read quite a few folks concerned with existential risk, and never saw this assumption before.
I was under the impression it came from Nick Bostrom. And other than some assumption or some Bostromian thought-experiment, where in the world did the phrase "trillions of descendants" even come from?
AI is not a question with an answer, and so will never be solved.
AI safety is not relevant to exploiting technology to ameliorate senescence.
Absent immediate focussed direct efforts in ameliorating senescence, by the time there is a level of AI technology which can make a dent in medical progress, at least a billion, probably many billion, people will have suffered and died who might otherwise have avoided that fate. And the loss of their productivity will impoverish the survivors - and so cause AI progress to suffer.
There are new companies working on applying AI to epigenetic data to extend human lifespan including this one "On a global basis and in aggregate, wet lab experiments in human biology produce one of the most important forms of data known to mankind, especially when it relates to epigenetics, genes, proteins, genomic pathways and environmental factors that can improve human health and extend human lifespan. We are a start-up with a focus on Life Sciences in the area of advanced artificial intelligence analytics connected to extending human healthspan and extending human lifespan." http://45.55.135.97/index-life.html
To expand upon this, it's not just publishing papers behind paywalls that is the problem. The entire system we use to advance humanities shared knowledge is broken. The Internet enables a dramatically more efficient model of scientific collaboration. The problem is academic science has been stuck with a reward system from the 17th century. And quite simply, it does not reward what we want it to reward anymore. However, there is a solution: science funders can distribute their money in a way that creates a very different set of incentives. Sadly, almost all science funders are very conservative organizations. YC Research can play a critical role in catalyzing a transition to an Internet-native model of research. My startup, Thinklab, wants to help. We're a service that helps science funders reward participation in a massively collaborative open online model of research.
Can we add some nuance here? It is not like in the last few decades there has been no new knowledge generated.
Just in my own field of biology, the amount of progress we have made in understanding how cells regulate themselves (20 year ago we were still teaching DNA->RNA->PROTEINS that do stuff), how immune systems function and can be used (stem cell therapy, cancer treatments, HIV drugs) and the flow of genetic information between organisms (the advance of (whole) genome sequencing into becoming a generic tool) has been mind-boggling.
I agree -- we're making great progress. Science IS working. My point is really just that since the advent of the internet, we've had the potential of making progress much faster.
We are without a doubt making progress, the question is, how much more progress could we be making with a better system to organize the thoughts and efforts of scientists around the world.
I think the computer science department has used internet really well with the open source projects. But, the same can't be said of most experimental sciences. In life sciences, we sometimes wait a few years from when an idea strikes a researcher to academic publication. An open-source collaborative approach could have immensely sped up progress. Not only that, the academic papers often describe only the experiments that succeeded. Anyone trying to reproduce the result will make the same mistakes as the original authors. What a blatant waste of human brain hours!
I agree completely about the system being broken and that innovation would be fantastic (I wrote a few paragraphs below, but in a very different direction). I'll take a close look at Thinklab, and it'd be great to see a dialogue forming here.
Great! I read your comments and have to admit I'm unfamiliar with what a "collaborative atlas" is or would be? But I certainly agree it's a big problem that scientists are only rewarded for publishing papers. There needs to be a way to reward scientists for sharing in smaller chunks -- even just an idea that might help out a peer.
Yes. I mean broken in the sense that the incentives of individual scientists are not aligned with what's in the best interests of science a as a whole. It's the reason scientists hoard knowledge and work in silos instead of sharing their work openly while collaborating over the internet. I've written more here: http://thinklab.com/blog/10-consequences-of-a-broken-scienti...
I wonder if scientific progress speeds up in wartime because scientists stop worrying about who gets credit and how things will affect their career and just learn to collaborate until victory is achieved.
Do we? Or do we need more realistic perspective about death? Why quantity over quality?
Seems there's a lot of quality improvements that could be made [1]. For instance, I read recently (can't recall source) that San Fransisco's homeless is the least happy in the world. Also that the top 1% in the world make over $35K a year, but there are far happier people than exist in middle America living in poverty [2]. Looking at populations that are happier - but in poverty - is perhaps an interesting study. Imagine a cross disciplinary study between (perhaps) techniques used in epidemiology/health informatics & anthropology / the humanities.
---
[1] Notice that cancer, heart-disease, Alzheimer, etc research also falls into this category - allowing people to live with less suffering.
[2] Yeah, 3 assertions, no sources, sorry - just my word & unreliable memory of something someone said in a book once folks :)
> Or do we need more realistic perspective about death?
We do: we need to realize how horrible it is that we treat it so casually, rather than as a tragedy. Perhaps then we'll do something about it.
> [1] Notice that cancer, heart-disease, Alzheimer, etc research also falls into this category - allowing people to live with less suffering.
All three of which (especially the first and last) have heavy ties to aging. It isn't just about living longer; it's about not aging. We're not talking about getting a few more decades like those in your 90s; we're talking about getting a few more centuries like those in your 30s-40s. (And hopefully that will buy us enough time for the next such improvement, and the next.)
Wanting to live forever has to be the apex of egoism and arrogance and all that is wrong with the human mind. Death is good because it leaves room for new people with new ideas.
Being against death reveals a profound misunderstanding about what life is about, which is: finiteness, urgency, novelty, creativity.
I think defining other people to be "profound[ly] misunderstanding what life is about" just because they do not believe the same as you when it comes to something so fundamental and personal as one's life is much closer to "the apex of egoism and arrogance and all that is wrong with the human mind".
We don't need room for new people with new ideas, room isn't the issue: as far as we know so far, the Universe is virtually infinite. When we come around to defeating death (it's bound to eventually happen) we'll start terraforming, doing generation-ships to reach even further, and the rate of growth of human knowledge will reach an even more unimaginably fast pace.
I suspect you're going to get hammered here but this is an incredibly important point. Aging and death forces the passing of the torch from the old generation to the new, and I think this is a crucial part of progress. A tragedy on par with death would be the rise of some oligarchical class of immortal technocrats, wielding power and influence for centuries on end.
I can see the job description now: "75+ years of relevant experience required." At the same time, allowing the Einsteins of society to continue their work indefinitely would provide a net increase in the rate of scientific advancement. Unfortunately climbing the corporate ladder would be nearly impossible as positions would almost never go vacant. It has both pros and cons, but allowing a larger fraction of the population to be in the workforce might just outweigh the negatives.
I don't think it would ever happen, but the only real counter I can think of to that is modifying our brains to have intermittent periods where they are more plastic than normal. Honestly, our current system works really really well... I still don't want to die though.
Hypothetical bogey men of an imaginary dystopia should not concern you nearly so much as the loss of productive life, physical and mental decay, and the reckless violence of the hopeless which surrounds you today.
A valid point, but why should the pessimistic outlook be any less likely than the utopian one? Are you sure that a major missing link between where we are now and a prosperous future for everyone is the fountain of youth? Sure, the HN'ers want the Einsteins and Feynmans of the world to live forever? But what about the Nixons, King Leopold II's, and Kim Kardashians of the world? Is it okay with you that they will be given first dibs at the antidote?
You're seeing it as the key to suffering, but its equally reasonable to view it as what it will likely become - an instrument of power.
>Death is good because it leaves room for new people with new ideas.
Congratulations: you've managed to point out one of the very, very few nice things about having your body slowly but unstoppably degrade for six decades starting in your mid-20s.
You can add more people without destroying the information contained in their predecessors. We are on the cusp of interstellar travel, once we are no longer bound to our singular space rock our population no longer needs to have limits.
Not everyone has to share your profound insight about the meaning and purpose of life.
This is a classic example of a limited rationalised mindset. Imagine how much more enriched people lives would be if they got to know their great grandparents and more. Death is a tragedy, there is nothing stopping you and others who share your beliefs from ending life at any moment they want. This is about options, think about the potential -- it would change society is so many ways.
Something I've been thinking about for a while is: what would it mean for peoples' willingness to save another person from a burning building, or swim to rescue someone drowning at sea, or donate a kidney, etc, if we all developed a sense of entitlement that we should be able to live forever?
When so many people at all levels of society are unable to enjoy the lives they have, I feel that life extension research is not something that should be a high priority for investment.
"Wanting to live forever has to be the apex of egoism and arrogance..."
I would gladly sacrifice my life to prevent the rest of the species suffering the physical agony, mental degeneration, poverty, frustration, and violence which result from pandemic senescence.
Your moralizing aesthetic judgements do not justify the misery of billions of people, or inaction in the face of its preventable imminence.
I think the key here is to introduce the concept slowly because saying "stop ageing" freaks people out. A lot of people have this idea of what "natural" means and stopping ageing just doesn't sit right with them. I think a better strategy would be to call it "eradicating diseases and higher quality of life towards the end of your life", both of which is really ageing research, but with a better spin.
We've been extending our lifespan for the last century through various technological advances. We are just continuing that trajectory, albeit at a faster pace.
> Is this just for rich people or do poor people get to live for centuries too?
"centuries" is far too short-term; "forever" is the goal, and "centuries" just buys enough time to get there. And of course that lifespan should be available to everyone. If the solution itself doesn't already imply a post-scarcity world, then it'll help that once you have a cure for aging, it's more economical to supply that to everyone than to treat the myriad complications that arise from aging. It's hard to get people to fund such a cure, but it seems far simpler to get funding for "we have a proven cure, let's get it to everyone".
> How would people remember everything?
In the short term, curing aging seems likely to help greatly with many degenerative mental disorders. In the long term, I expect people will find increasingly successful ways to augment their own capabilities. I doubt I'll have the same brain structure a hundred thousand years from now that I do today, if only because I expect we'll eventually run out of effective ways to debug biology in-place.
> How would you spend all that time? What would everyone do?
What wouldn't everyone do? I don't have a finite "bucket list"; my list contains everything positive I can possibly imagine, with a sort order applied, and it will always grow faster than it shrinks. Quite apart from the rest of the universe, other people provide an unbounded source of novelty.
Do you think people thousands of years ago would have asked such questions about a 120-year lifespan, and think it sounded too long?
If you lived for ten thousand years by default, would you think it too long and decide to die at 120? How about if you lived indefinitely?
> > Is this just for rich people or do poor people get to live for centuries too?
> "centuries" is far too short-term; "forever" is the goal, and "centuries" just buys enough time to get there. And of course that lifespan should be available to everyone. If the solution itself doesn't already imply a post-scarcity world, then it'll help that once you have a cure for aging, it's more economical to supply that to everyone than to treat the myriad complications that arise from aging. It's hard to get people to fund such a cure, but it seems far simpler to get funding for "we have a proven cure, let's get it to everyone".
You do know that every culture, since the beginning of time, has thought that they would live to be immortal in their lifetime. What makes you think you're different? Sure, you might have a better chance, but it's still a slim chance. Pining for a cure to aging rather than living your life is a sad way to spend something as short and precious as your life.
> > How would people remember everything?
> In the short term, curing aging seems likely to help greatly with many degenerative mental disorders. In the long term, I expect people will find increasingly successful ways to augment their own capabilities. I doubt I'll have the same brain structure a hundred thousand years from now that I do today, if only because I expect we'll eventually run out of effective ways to debug biology in-place.
> > How would you spend all that time? What would everyone do?
> What wouldn't everyone do? I don't have a finite "bucket list"; my list contains everything positive I can possibly imagine, with a sort order applied, and it will always grow faster than it shrinks. Quite apart from the rest of the universe, other people provide an unbounded source of novelty.
I don't think so. Wanting to live forever is a childish thing, it's a fear of death mixed with a type of greed. Actually living forever would be a greater hell than anything I can imagine. After the first few hundred years, when you realise that it's never going to end, after you've read every book and done all you've wanted to do, what then?
> Do you think people thousands of years ago would have asked such questions about a 120-year lifespan, and think it sounded too long?
No, because the maximum lifespan of a human hasn't fundamentally changed. The only difference is that more people are living to old age.
> You do know that every culture, since the beginning of time, has thought that they would live to be immortal in their lifetime. What makes you think you're different? Sure, you might have a better chance, but it's still a slim chance.
A slim chance is infinitely better than no chance at all. How many cultures had even the remotest chance of actually doing something about it?
And no, I'm not certain it'll happen in my lifetime. I'd say that I'd be sad and disappointed if that didn't happen, but really if that doesn't happen then I won't be anything at all. But I still consider it by far the most important problem that could possibly be solved, and worth putting incredible effort and resources towards. I also consider it worth advocating, to encourage others to push for the same goal, or at the very least discourage others from perpetuating arguments that shut down such efforts.
> Actually living forever would be a greater hell than anything I can imagine.
Then don't. But I'd suggest trying it first, or significantly expanding your imagination.
> After the first few hundred years, when you realise that it's never going to end, after you've read every book and done all you've wanted to do, what then?
With all the imagination you can bring to bear, and a universe full of possibilities, you can only think of a few hundred years worth of things to do before you'd not only get bored but get so bored you'd long for death?
To give even a minimal lower bound based on your own comment, books and stories are already being written today faster than they can be read. And that's only one of myriad possibilities. A few minutes imagination can easily produce far more interesting ones.
Do you currently get bored with life and want it to end? If not, then why do you expect that to change in only a few hundred years?
> > You do know that every culture, since the beginning of time, has thought that they would live to be immortal in their lifetime. What makes you think you're different? Sure, you might have a better chance, but it's still a slim chance.
> A slim chance is infinitely better than no chance at all. How many cultures had even the remotest chance of actually doing something about it?
None, and we're no different. Why would we be?
> And no, I'm not certain it'll happen in my lifetime. I'd say that I'd be sad and disappointed if that didn't happen,
Well, no. You'll be dead. Dead people aren't disappointed.
> I also consider it worth advocating, to encourage others to push for the same goal, or at the very least discourage others from perpetuating arguments that shut down such efforts.
I don't agree that it's worth advocating. Yes, curing diseases and other such things is a worthwhile goal. But the goal of living forever is just selfish. Note that children couldn't exist in a world where people live forever (otherwise we'd run out of resources even faster than we are now).
> > Actually living forever would be a greater hell than anything I can imagine.
> Then don't. But I'd suggest trying it first, or significantly expanding your imagination.
I have a very vivid imagination. That's how I came to that conclusion. Also, how can I "try it"?
> > After the first few hundred years, when you realise that it's never going to end, after you've read every book and done all you've wanted to do, what then?
> With all the imagination you can bring to bear, and a universe full of possibilities, you can only think of a few hundred years worth of things to do before you'd not only get bored but get so bored you'd long for death?
Yes. Why do you think any differently? Evolution has placed a cap on our lifespan, because living longer than that wasn't better for our species. Have you considered that? Why do you think we'd be able to contribute anything meaningful to the world after we are 80 years old? Aside from helping younger generations (which don't exist in your world), we have nothing left to do.
More than 98% of people who ever lived are dead. It's incredibly arrogant (and akin to a tantruming child) to assume that you will survive.
> To give even a minimal lower bound based on your own comment, books and stories are already being written today faster than they can be read. And that's only one of myriad possibilities. A few minutes imagination can easily produce far more interesting ones.
I'm a scientist (as well as programmer). So I can imagine spending several hundred years trying to solve all of the scientific problems that exist. But after a few hundred years, I'll definitely get bored of that. So, I'll have to move on to something else.
> Do you currently get bored with life and want it to end? If not, then why do you expect that to change in only a few hundred years?
No. But that doesn't mean I won't get bored of it eventually. Not to mention that I can see myself being bored with <things I'm interested in now>, and no doubt I'll eventually get bored of all of the things that make me a productive member of society. At that point, I'm a drain on the world's resources. What benefit is there to keeping me alive?
If we cure aging and live for centuries we will still be subject to accidents while we are trying to work out how to live without a body at all. I would be interested to know how likely it is to get maimed or killed during this extended period and how it would affect the psychology of a person. I assume it would be a pretty frightened bunch not leaving home much.
One of several rough estimates I've seen suggests that if you eliminated all biological causes of death, and left only accident and similar, that would bring the average lifespan to around a thousand years. Producing such an estimate seems straightforward enough, given statistics about causes of death.
As for how that might affect psychology: honestly, I'd love to see how the world might change if people adopted drastically different attitudes towards death, but I don't think it likely that this would produce far more risk aversion than we have today. Not least of which because I see no signs of tobacco companies going out of business.
For myself personally, I'm not going to stop leaving the house, but I certainly have no intention of taking up hobbies like hang gliding or motorcycle racing, any more than I plan to start smoking. (Side note: for an interesting measure of risk, see https://en.wikipedia.org/wiki/Micromort) I don't think that level of risk aversion qualifies as "a pretty frightened bunch not leaving home much".
Living for centuries would be absolutely awful. Death is necessary. It's the world's chnge-agent. Think about how slow the rates of social progress would be without death.
Also the fact that life is only a fleeting moment in time is beautiful and again, I believe necessary.
Therapies after the SENS model will be mass-produced infusions (gene therapies, small molecule drugs, etc) coupled with some personalized stuff based on growing cells from a cell sample. It will be the same treatment for everyone, and subject to economies of scale that are the same as for today's drugs. Most will cost a few dollars per treatment after being in circulation for a few decades, the same as mass produced generic drugs today, and 30 years further into autologous stem cell therapies, those also will cost a lot less than the few thousand that they do now.
People have done the modeling on what radical life span does to population growth. The answer is a lot less than you think. See: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3192186/ "For example, we applied the cohort-component method of population projections to 2005 Swedish population for several scenarios of life extension and a fertility schedule observed in 2005. Even for very long 100-year projection horizon, with the most radical life extension scenario (assuming no aging at all after age 60), the total population increases by 22% only (from 9.1 to 11.0 million)."
Most of the objections of the type that you voice can be answered by moving you to 1850 and making the same objections to the progress of medicine, and then seeing that they are obviously false concerns given where we are today. You could argue in 1920 that the same objections exist to building treatments that prevent people dying from heart disease: isn't it just for rich people, won't there be too many people if the old don't die on time, what would they do with all those extra years, isn't the present length of life just right, etc, etc.
Why not? Veterinaries would love novel approaches. Plus the pre-research is always carried in animals.
> How would people remember everything?
You should ask Google's Director of Engineering (Ray Kurzweil) about that. Hint: within 25-30 years your brain could have x1000 capacity via cloud computing (mobile + brain-cloud interfaces).
> Is this just for rich people or do poor people get to live for centuries too?
Your questions are quite common and have been answered hundreds of times in any of the SENS presentations. Answer: humanity as a whole will benefit from this research.
> Haven't done the math but if 8 billion people lived for say 500 years each, plus new arrivals constantly, wouldn't we accumulate rather?
The fact that we live twice as long, and we have duplicated the human lifespan in the last 100 years (in the western world), but now we have less natality tells you something? Overpopulation comes from the 3rd world and decreases as a country starts to develop...
You must be young. That's what people were saying 50 years ago. When I was in high school the reason I was told we had to learn square dancing was because in the future we would have so much leisure time.
I remember when I first deconverted. I was horrified that there was no heaven, and that my loved ones weren't waiting for me. That they were just gone forever. They didn't exist. They weren't happy, and I would never see them again.
I see many atheists carry on religious memes after they deconvert. Which is only natural, religion isn't just a single idea, it's deeply ingrained into our culture. One of those memes is that death is natural and ok. Which, if you believe in religion, it absolutely is. It's the way things are supposed to be according to a trustworthy higher power. There is a life afterwards that makes it all ok. People don't really "die", they just go to a different place.
I don't see anything ok about true death. It's natural, sure, but there's no law that says nature has to be good. We are the product of an uncaring universe and plenty of bad things exist that shouldn't. Death is one of those things.
So I came to the idea that if Heaven didn't really exist, we should build it. I was quite young at the time and influenced by science fiction. I thought that maybe future humans could invent time travel and come back in time and save people who died.
I've since come to the conclusion that time travel is very unlikely to be physically possible in our universe. But other advanced technologies, like curing aging, are certainly not. Perhaps even within our reach. And if backwards time travel might be impossible, we have invented a way to do forwards time travel with cryonics.
> One of those memes is that death is natural and ok. Which, if you believe in religion, it absolutely is.
To be clear, this is pretty untrue of Christianity, the dominant religion in Western civilization, so I'm constantly puzzled about this claim. (HPMOR makes the same claim.) The central story of Christianity is that God takes human form, dies, and cannot stay dead. If it were just "Jesus died for our sins", there would be no need for a resurrection story; his followers would have had way more credibility saying merely that he was killed, which is far more objectively verifiable.
From 1 Corinthians: "The last enemy to be destroyed is death" (which, incidentally, comes up in canon HP), and, quoting the prophets: "When the perishable has been clothed with the imperishable, and the mortal with immortality, then the saying that is written will come true: 'Death has been swallowed up in victory.' 'Where, O death, is your victory? Where, O death, is your sting?'" From Revelation, also quoting Isaiah: "He will wipe every tear from their eyes. There will be no more death or mourning or crying or pain, for the old order of things has passed away." From a 1500-year-old hymn: "Christ is risen from the dead, trampling down death by death."
Death is an enemy; death is not good, and should not exist. Perhaps other religions disagree, but Christians and atheists should be able to agree on this basic fact.
I don't know what scripture says, all I know is that actual Christians believe that their loved ones are in Heaven and that humans have immortal souls. There are a lot of memes in our culture about death from religion. Just go to a Christian funeral and listen to the pastor talk about Death is natural and ok. Of course they still think death is sad, but the religion provides comfort and meaning to it.
I think death is probably a selective evolutionary advantage for a species.
A species is basically self perpetuating genetic code stuck somewhere in an evolutionary local maxima. In our case there may be an advantage in adaptability by having individuals die and genetic code mix/change more frequently.
This still sucks and if we could figure out how the brain actually worked to prevent it that would be cool, but we'd start getting into identity issues (what makes you you? how does consciousness actually work?).
Sleep is interesting to think about here since it likely has something to do with re-calculating weights in a neural net, garbage collection, persisting memories etc. Though we don't consider going to sleep dying and waking up new because of the sense of continuity. If a copy of you was made and the original then killed though we wouldn't like that.
Sense of purpose difficulties still exist in our relatively brief lifespans, but at least we get an easy out with the biological one - we'd have to think more about this with death out of the way. Also the drive to have sex and reproduce would probably still be around since it's core to our existence, not sure how we deal with that.
Though I'd still prefer a star trek future where we're colonizing the galaxy and the goal is to learn.
The evolution selects for death idea is a hypothesis, but I don't' think it's very strong.
Evolution works at the level of genes. Genes that cause more copies of themselves to spread will spread. Even to the detriment of the group or the species.
An organism that lives longer will have time to have more offspring, and those offspring will carry it's longer living genes. An organism that lives shorter will have fewer offspring, and so fewer organisms will carry it's genes. Very rapidly these genes would be selected for or against.
A simple alternative explanation is just that evolution doesn't care. 90% of organisms will die of other causes long before they reach old age. Evolution will end up devoting almost all of it's resources on improving survival rates there, and none on fighting aging.
>This still sucks and if we could figure out how the brain actually worked to prevent it that would be cool, but we'd start getting into identity issues (what makes you you? how does consciousness actually work?).
You can live for a thousand years and you will still procrastinate and you will still not want to die. An extended lifespan won't help those issues. You need to get over them whether you live 50 years or 500.
> You can live for a thousand years and you will still procrastinate and you will still not want to die.
Of course not! Why should you ever want to die? Why should you ever "get over" that?
It's a bug. We haven't fixed it yet. It should not be romanticized, and it should most certainly not be treated as inevitable. It's an abomination we have yet to eliminate.
There is a pine tree that is almost 5,000 years old, and a number of plants "reproduce" by spreading out their root system where what we see as a multitude of plants is essentially one long-lived organism under the surface.
For humans (or at least for me), I think the main concern is really more of continuity of experience than maintaining a particular body configuration indefinitely. I.e. I would like to run into an old friend in 10,000 years and reminisce about that time we did something hilarious. It wouldn't be very satisfying to undergo a procedure that resets my body but wipes out my memories in the process.
It wouldn't be very satisfying to undergo a procedure that resets my body but wipes out my memories in the process.
We whipe out memories every day. Your memory of your childhood is a half imagined highlight reel of moments. If you lived 10,000 years, you might only have a couple moments left from this century.
Do you think of cancer as inevitable? Or do you think of it as something humans could cure someday?
Do you think of degenerative mental disorders as inevitable? Or do you think humans can cure them someday?
I don't think of death as inevitable. I think of it as something humans can cure someday. Thinking of it as inevitable means giving up on all possible solutions.
I think you've used an unconventional definition of "death" here, distinct from the one thus far used in this discussion.
Lack of change seems closer to death than change does.
But in any case, whatever definition of death you've used doesn't seem to relate in any way to the ones that people in this discussion have advocated eliminating.
If you 'survive' (say every cell is replaced) then what is "you"? I'd say it's more likely to be a set of linguistic & cultural norms than it is to be anything fundamental from the view point of physics.
So that raises the point: is 'you' conceptual or is there anything fundamental? It makes sense for evolution to have selected for organisms to believe there was, no matter what, something fundamental, existant & worth preserving about them, but really - we're this group of cells & the plants we ate for breakfast (and the efforts of the farmer of those plants and so on)
Of course, as a "manner of speaking" (that is, in common every day culturally contingent world), all this is bogus.
But from a fundamental, naturalistic point of view, I don't see anywhere else to go :) It's also a very parsimonious view as you get to explain a lot of stuff with less entities, etc :)
In my own life there are several things I'd like to do that will take somewhere between 5-10 years each. I only have 2-3 of those spans left and 100 of those things I want to do.
Wanting more lifespan is a different thing. The claim that most people want to live longer is not as controversial as the claim that most people need to live longer.
If you will certainly die for the lack of some thing, then you can truly say that you need that thing.
Senescence and death costs the species and the planet a horrible price in lost productivity, social conflict, medical costs, human desperation and suffering. Even if you don't want to live beyond some fixed span, if you want to impose suffering and death upon others...well, I don't even...
Agreed. With overpopulation already a problem I'm not sure that living longer is a pressing need for the world. Better to help people make the most of the time they have allotted on this planet.
Thank you for sharing "Don't Panic", watching that is really an hour well-spent.
Since permanent growth comes from fertility and as "Don't Panic" shows we're getting to a sustainable average of 2 child per woman, augmenting our lifespan will only increase our maximum population, not our population growth.
However I think YC research would be better spent trying to find better ways to fight climate change, world stability and/or lift billions out of poverty. This way if we ever augment lifespans in a significant way it will be available to more people and in a healthier world. Also lifting billions out of great poverty will have the added benefit of allowing more minds to work on issues like increasing life expectancy.
Also, we should really focus on getting people on board with the life-preserving tech that currently exists (medicine, water treatment, doctor training, etc) before developing new (and potentially expensive) tech that will only widen the gap between what the rich can do that the poor can only dream of.
Indeed. From a biological standpoint, it's important to realize that an indefinite lifespan, as a life history trait [1], does not confer evolutionary fitness. The fact is that organisms do not appear to evolve to a maximal lifespan — they evolve to an optimal lifespan. Humans are already considered to be K-selected [2], and increasing lifespan appears to necessarily decrease fitness in other areas, as has been seen in worms [3] and flies [4] (e.g. reproductive fitness decreases and age of maturity increases, when lifespan increases).
More importantly, "near-infinite" lifespans reduce the efficacy of natural selection — longer-living organisms evolve at a slower rate. While, perhaps, unappealing, the culling of species members well past reproductive age is necessary to allow for more resources and potential for competition and selection in their offspring. And humans are certainly still undergoing natural selection in many arenas; just look at sickle cell trait [5] or or CCR5-Δ32 [6].
These are all arguments that apply to the house you live in, the blankets on your bed, the medicine you use. You are absolutely fine with all of those - you don't live in a cave, shun treatment for infectious disease, and wear furs. In fact I would go so far as to say that you probably think that modern technology is a good thing, and you are better off for it. Yet that already greatly manipulates your proposed optimal condition for the human species. So why does treating aging as a medical condition bring out this view? This seems like a very selective appeal to nature.
It may not have been expressed the most clearly, but the intention was not quite as you describe. Shelter, clothing, and modern medicine all confer enhanced fitness — I was attempting to suggest that increased lifespan does not.
Reminds me of Seneca's De Brevitate Vitae[1], an essay about (as the title suggests) the shortness of life. He argues that what matters is not how long one lives, but what he accomplishes and/or how "well" he lives.
It is important to recognize that arguments for quality over quantity are written from the perspective of operating within an absolute and unchangeable limit, and an absolute and largely unchangeable high risk of unexpected death before that limit due to infectious disease. We've done a lot to reduce the latter, and now the former is up for grabs as well. This is not the same landscape for the human condition as occupied by Seneca and his peers.
We can work towards both quality and quantity. That is the point of technology, to make everything better, to remove limits, to enable the human condition to be more than it was before, to give us choice where there was no choice. The choice to fly, to talk to distant people, to be healthy, to live rather than to die before we want to.
it's not quantity over quality, they are directly proportional to each other, maybe, if meassured in absolute terms. There's an optimum and quantity is hampered by lack of quality.
I’d have said AI safety, but it’s already on your list. Excellent.
But, why? Nothing about "AI safety" is remotely useful. It would be like researching "alcohol safety" (something everybody has access to and nobody can restrict) or researching how ants can influence "human safety" (tiny things trying to influence massive things). It's nonsense no matter which way you think about it unless you just want to write SF novels.
Open-source infrastructure.
Sadly, we don't need "research" to study how it works, we need direct payments to contributors. As a "startup culture" we need to stop the trend of billion dollar companies using free open source labor to reduce corporate costs and boost corporate profits while paying highly skilled professional volunteers nothing.
Yudkowsky Ambition
A key part of _research_ is you need to have a plan of attack. Saying "We're going to fund research into antigravity, time travel, and zero point energy" sure sounds like fun, but it's an absolute waste of time.
> boost corporate profits while paying highly skilled professional volunteers nothing.
This is a problem of ancient outdated patent and copyright law, not scientific innovation. Its in the field of politics since the industry is entirely built on state enforced IP legislation.
But open source code is all about "no patent restrictions" and free licensing even with retained copyright (or even outright copyright assignment/dumb CLAs).
Iterating those concepts don't seem to fix the issue of Twitter running 50,000 copies of an application developed by 3 open source volunteer people and not paying those 3 people anything.
The issue is that with copyright and patent law the way it is written software is heavily biased towards proprietary development for profit, by restricting what the user can do with the software they run. It is also an issue of users not caring that they are being given the equivalent of a car with a sealed engine compartment. The average consumer can understand why that is bad, even if they themselves would never open up the engine, because it would mean they could not let their local mechanic fix their car, they would have to submit to the manufacturer at whatever price they want to charge because there is no competition anymore.
They just don't correlate that to proprietary software at all, which is understandable because most people cannot internalize concepts related to information transfer, scarcity, transformation, or preservation. They exist in the real world and more easily relate to the physical and ignore the implications of ignorance in all that is not presented to them in that medium.
That still remains a political problem, one with your system of economics. I guess since YC is funding that UBI research it may also be relevant to research alternatives to US IP law that can optimize for productivity and social utility of the results, because we are obviously not close to a maxima right now.
If i could pick just one, it would be academic publishing. It's an ancient system that locks in tons of useful empirical data and rots the roots of science with its publish-or-perish/impact factor lunacy. It's also a sector (digital publishing) which YC knows well and can easily make a great contribution. This can be a game changer, as it can alter the way public funding is allocated.
I would say along the lines of academic publishing, there is also a problem with people publishing misleading or plain wrong press releases. I think there needs to be a better way of evaluating the reliability of research, and a way to possibly improve the peer review process. This may mean that research has to be more open with the data that they collect. I think there is definitely some work to be done here.
Mental health - specifically in a future where we may be increasingly more connected virtually but increasingly less connected physically. In one or two decades when the average urbanite is serviced by on-demand autonomous vehicles and drones, works from home (or not at all), consumes news/media in a personalized subreddit-like echo chamber interspersed with cleverly integrated native advertising, and spends 10+ hours/week in VR, what are the implications for their mental health? As physical community and socialization are often pointed to as the greatest predictors of happiness, how do we best translate that deep seated human need into the future? As medical science continues to improve treatments for cancer, worn out joints, failing organs, and other physical ailments, I think the mind will eventually emerge as ''the final frontier'' of health care and well being.
We still don't know what long-term exposure to highly condensed, highly customized information does to our brains and social systems as a whole.
For example, there are signs that social network users are actually unhappy - in other words - are experiencing all kinds of mental health issues.
It is well known that programmers experience "burn outs", which, from personal experience, is a terrible and debilitating state sometimes even leading to suicide.
There is addiction to gaming, chat, porn, social networks.
Mobile phones have exposed everyone to the infinite stream of information/communication and I think this will lead to many more mental health issues.
I think we are not yet prepared for the kind of impact that current and future tech will have on our minds, so this would be a worthy field to invest in now.
Came here to say the same. Last night I was hanging out with my 3-year-old nephew and wondering how different his life will be, especially with all the advancements like VR coming our way. I feel lucky to be old enough to remember what life was like before the internet.
The idea of people spending 10+ hours in VR per week scares me, but it's probably pretty similar to video game and smartphone usage. Maybe that would be a good place to start research.
A little over a month ago I started working on forming new habits, severely limiting use of network tools. I now only check email/sms/etc twice per day. At 6pm I put all technology away. I'm asleep by 9:30pm, awake at 5:30am, and try not to look at any network tools again until 10am. I'm considerably happier and more productive now. It's a tough habit to maintain and I'm pretty sure a few of my friends think I'm nuts.
This is something I'm trying to address with my startup, Krewe: https://www.gokrewe.com/about. I want to make it easy for people to quickly form strong, lasting friendships and relationships and become connected with their local community. It gives people access to a small social group in their neighborhood and gradually allows them to expand that group. Since everyone lives so close to each other, it makes it possible to get together often, even everyday if they want, without having to deal with extensive planning or traffic.
If YC or anyone else wants to fund me, I'd be fine with that.
Everyone knows it's broken,not effective anymore and literally a waste of time and money. More so if you had the oppurtunity to experience what education means in third world countries.Spoiler alert : it's a joke that will make you cry.
It needs to be fixed as quickly as possible and research needs to be done for alternatives to help people in the process of learning.Online education is a good start, but we need something more effective.
By not fixing education, we are creating generations of people with just pieces of paper(called degrees), having no real knowledge or even the thirst for knowledge. Maybe I am not vocalizing my thoughts well enough, but you just need to look at the state of education in third world countries to understand my depression about education.
This. In keeping with Gates' philosophy, when working effectively with non-governmental levels of capital it's critical to make sure you leverage as much as possible to effect real change. Things that play out across a long timespan (relative to a human lifespan) are an excellent opportunity to accomplish that.
Education: We've had teacher-run schools. We've more recently had test/metric run schools. Let's (in the US) finally try and effectively fuse those two into better outcomes per dollar for kids in public schools. My personal pet peeve: develop a modeling technique that predicts an "average" educational outcome for a student given his or her family's socioeconomic status and other inputs. Then at least we have a basis for a fair measure of teacher success -- no teacher can turn every child who grows up without a single book in the home into Albert Einstein or Marie Curie.
Economics: Post-scarcity minimum / basic income, as mentioned.
Democracy: Follow on and unify the various open-* voter information efforts (example: http://votesmart.org/ ) and find the best and most unbiased way to communicate that. Democracy only works with an informed electorate, and we can do much better than throwing our hands up and letting the most ad buys have the strongest influence.
Nutrition: There's a huge amount of bogus advice out there. Synthesizing this into actionable, individualized suggestions and helping people eat healthier by unifying food and recipes with budget. Eating healthier for one day does you no great help -- eating slightly better for your entire life does.
Education isn't ideal now, but I don't think online education is a good start or even a viable model.
Training more and better teachers is the way to go, especially in "third world countries". Not only is there big issues with motivation and the whole social/interactive/distant aspect to moocs; what do you do when you don't have internet, or your device breaks? Now your education depends on a machine that costs a couple hundred bucks, instead of a notebook, pencil and early morning.
Idk, I don't see online education as any more than just a sometimes-useful alternative.
> I don't think online education is a good start or even a viable model
Let's suppose you open-sourced education, so rather than it being controlled by an elite and passed down to the masses in varying degrees depending on how much money the recipient has, it's controlled by everyone. People can design their own courses in the same way you make a Linux distro or other open source project. Courses could be stored somewhere like Github and downloaded, forked and merged like we currently do with software. Suppose people taking those courses could contact the developers or their helpers if they don't understand something, or just want to discuss it -- email, Skype, Slack, etc. So many people want to learn, and so many others (like me :) would be happy to help them if there were the right structures and support systems to make it possible.
All these things could come together to become something far more than just "online courses", just as open source software is so much more than software you don't have to pay for. Imagine we're at the Apple II stage of online education, but the iPad version (30 years down the road) will be unimaginably cool and powerful. Let's just imagine a world where you expect people to be educated to a decent level, just as we increasingly expect a base level of computer literacy. You're right that the infrastructure is not there at the moment, but let's not get hung up on that -- it will spread and increase, so what good things do we want to follow it?
If online education stays where it is, you're right -- it's not good enough. But, respectfully, I hope you're wrong, and we're seeing the start of something extraordinary, the next phase in the democratisation and distribution of knowledge. I hope universities start to look more and more like museums, and education becomes a lifelong endeavour for everyone: learners, teachers, and all points in between.
Personally I think this would be a tremendous thing for PG&Co to research. Hopefully, we're just getting started :)
I believe education will find a natural middle-ground.
In many instances the benefit of having an in-person instructor has much to do with the social interaction/interpersonal support that they can give to a struggling/minimally motivated student. That said, motivated learners don't always need that kind of personal interaction and would benefit from self-guided learning opportunities. This is obvious, but is hard to manage well.
In the end, I believe students will learn along a fluid spectrum from 'one-on-one' to 'fully distributed online learning' - both of which will heavily lean on better tools.
There are tens of thousands of lessons, activities, individual interventions and curriculum modifications created every day by instructors in this country - but all of that knowledge and all those artifacts are effectively lost (as almost nothing gets distributed) and the same creative labor is repeated over and over.
So - we should be looking for better ways of mediating in-person instruction to both improve that kind of instruction and as an engine for the creation of universally accessible learning materials. Students who have the ability/proximity to a physical schools get the benefits of personal interaction, while students who cannot get to school (or are underserved by that school) can get the benefits of time shifted lessons/activities/etc. from anywhere in the world.
^ That sounds viable. I want to point out though, that I hope this doesn't mean states will stop trying to get more schools closer to people, or more teachers trained. As good as one-to-one, "personalised" instruction can be, students gain a lot from learning and working together with other students in a classroom.
Tech should be a tool, and not a replacement for teachers IMHO.
Absolutely - and I make these suggestions as a high school teacher. I also have downriver concerns about mass, compulsory data collection on the cognitive development of our youth (though that is not a concern exclusive to education).
It's teachers that are not remotely viable in the developing world.
I read recently that the teacher failure rate in Maharashtra is about 15% - i.e., on a given day, 15% of Maharashtrian teachers just don't show up to work, and there is a famous case of an MP teacher who's been absent for 23 years. [1]
Further, even if the teacher shows up, the student may not be able to. Consider the "town" in this pic: https://imgur.com/mjWcEZZ Hiring a teacher for the 5-10 kids who live there is unaffordable, so the kids there need to climb the "mountains" [2] whenever they want to attend the closest school. Of course, the kids don't actually do that every day; work and family obligations often interfere.
A 5-10k s mobile phone with a (worst case) 250rs data plan works far better than 85% of the time, even in that rural town. It works on the student's schedule, so if a horde of visitors require the student to spend the day in the family restaurant, they can catch up at night.
The real thing which would help a lot is better devices. E.g., something closer to a kindle in terms of battery life an readability.
Education is about much more than just a transmission of knowledge, however. IMHO efforts are better spent training better teachers, than making and buying more devices.
A good teacher will inspire kids, will get them to work together, will be a person to go to. I do believe in the motivation of students, but sometimes you need someone to spark that motivation in the first place.
There will always be bad teachers, and outrageous stories about them making way around the internet. You hear about good teachers much less often, but their impact in a community can be much, much greater. Teachers are viable; they worked for the developed world, back when they still weren't "the developed world".
A good teacher will inspire kids, will get them to work together, will be a person to go to.
What evidence do you have that even a fraction of teachers do this? And what evidence do you have that teachers "worked for the developed world" (i.e., had some measurable beneficial effect)?
There will always be bad teachers, and outrageous stories about them making way around the internet.
The stats suggest lots of them are bad. Rather than more teachers, maybe what we really need is a device to ensure that teachers actually come to work and the political will to stop paying them if they don't.
Online education is a great way to democratise access to the very best teachers/classes/content and I see it as a great way for learners to discover new topics/subjects quickly without needing to enroll in any physical classes.
I don't think we will reach the point where we have great teachers everywhere. That is just not practical and too idealistic. That's exactly why we need to start innovating in the field of education to try and make it a more meaningful experience and one which positively impacts society.
I see online education as one of the pieces of the education puzzle.
You can't just assume costs away. Teachers are very expensive and the main cost of education. I don't know if technology is cheaper yet, but it's worth considering.
Aubrey de Grey's SENS approach to creating and maintaining a state of negligible senescence in humans.
Every advanced society faces an aging population. This population will put massive strains on healthcare expenses. It could also strain the economy as the elder population collects public sector pensions and slowly depletes its savings (i.e. retirement funds). The latter effect could be mitigated by automation.
Paradoxically our success in treating diseases that kill us now will merely make us victims of potentially more terrible diseases in the future. Look at the experiences of centenarians in the last decade:
Alzheimer's will become increasingly common as will cognitive decline in general. What happens to all those people who don't die of cancer or heart disease? They will often succumb to slower acting chronic diseases involving long-term states of suffering and mental anguish.
The way out of this mess is to treat aging itself, not just the diseases that currently kill us. SENS targets the underlying mechanisms of aging.
Research into negligible senescence can also lay the foundation for potentially profitable therapeutics. Unlike the disease-centric model the total addressable market is nearly everyone.
I'm not sure it makes sense to put effort into these kinds of programs. As an example, a specific approach SENS wants to develop is lengthening telomeres in your body. But scientists are not even sure that shortening telomeres contribute significantly to aging-related diseases.
Derek Lowe (who knows drug development better than any of us) made a very interesting comment recently on the cancer research "moonshot" funding presented in the State of the Union speech:
"""
Trying to cure cancer in this way would be like trying to go to the moon without really knowing how rocket engines actually work, without being quite sure if Newton’s laws of motion would hold up, and with some real uncertainty in the position of the moon.
"""
The disparity between what we know today, and what we would have to know to "cure" cancer, is quite unfathomable to us computer hackers.
Perhaps ironically, lengthening telomeres such as via telomerase therapies is one of the things that a lot of other researchers are hot on and is not actually on the SENS agenda.
Since those other researchers are definitely advocating progress towards the use of telomerase therapies in humans, and it is inarguably the case that telomerase gene therapy extends life modestly in mice, probably by stimulating stem cell activity, your point still seems incoherent. See for example this position piece by Maria Blasco: http://dx.doi.org/10.12688/f1000research.7020.1
From my position telomere length looks a lot a measure of aging rather than a cause, and telomerase therapies are something I'd consider risky at this point in humans - our telomere dynamics and telomerase setup is very different from that of mice, and I don't think it is safe to assume that slowing of aging and induced regeneration in mice without cancer risk is necessarily going to happen in humans. You don't know until you try, of course, and there is a contingent that will be trying. Note that at least one human has already disagreed with that assessment and had telomerase gene therapy, the CEO of BioViva.
From the SENS perspective, telomere length is something that will take care of itself if you create rejuvenation by repairing the causes of aging. Average telomere length in tissues is a product of stem cell activity and cell turnover rate, and aging diminishes the former, and that is a reaction to rising levels of damage. Get rid of the damage and stem cells should get back to work because the signaling environment will revert back to how it is in youth.
<From my position telomere length looks a lot a measure of aging rather than a cause>
Sure, so they need to just keep in mind that the goal is not to lengthen (or protect) telomeres for their own sake. But telomeres could at least help serve as a metric or proxy for therapies that do slow or arrest aging overall.
Large Scale Medical Data Mining research, similar to OpenAI.
Specifically Computational Healthcare a Search and Aggregation Engine for Medical Records & Claims.
We believe that this is a classic Software eating the world situation and the time is perfect for it.
We have access to almost 130 Million de-identified medical records from approximately 36 Million patients (~10% of US population) this includes all Inpatient, ED, Ambulatory Surgery records between 2006-2011 from California. To put simply if you lived in California and went to a hospital there is 95.9% chance that we have your data. This data has been available for quite some time but its use has been hindered due to lack of good software. The data has led to significant research, e.g. my collaborator (not me) published a paper showing risk of strokes following pregnancies in New England Journal of Medicine last year.
At Cornell Tech & Weill Cornell Medical College, we have developed a Search and Aggregation engine that will revolutionize how researchers and physicians use this data. Imagine your mother with Leukemia in Remission just got admitted for Pneumonia. With our software, the Physician will be quickly able to asses likelihood of this occurring and rule out any confounding adverse events. Or consider that there is a rare combination of diagnosis e.g. Graves Disease and Clotting disorder that is indicative of a unique genetic mutation likely to offer novel insight into disease process. With our software questions like these can be answered within second, Today & Right now.
The Data, Legal structure and fully functional prototype are available right now. We were counting on support from AHRQ, but sadly the agency has run into trouble due significant budget cuts.
I agree this sounds a great candidate for YC. I hope you are looking to take this global. The cost/benefit in health outcomes would be enormous.
Most importantly, I think a system like this must be non-profit and freely available for global usage. Privacy issues can be solved with careful ETL process. Ideally a single global platform...although political realities may force multiple instances.
Regardless of implementation details - this seems to fit the YC Research charter in terms of medium/long horizons and big 'change the world' payoffs.
Thanks, Alex.
I agree that such system should be global / non-profit & free. We have also studied privacy issues surround such system in detail. A large motivation is that regardless of the privacy preserving technique employed the system is a huge improvement over current practices which involve sending out entire data to each individual research group.
Today a physician suspecting novel association e.g. Adverse event particular to a co-morbid condition. Usually has to wait multiple weeks for Ordering data from US government, Developing SAS scripts and finally conducting the study. Wrth our stack the underlying question can be answered within seconds. With enough legal permissions we can modify it to deliver only the required data, for further statistical analysis. Sure such a system might not be completely public, but there is nothing that prevents us from giving access to say all board certified physicians globally.
>"We have access to almost 130 Million de-identified medical records from approximately 36 Million patients (~10% of US population) this includes all Inpatient, ED, Ambulatory Surgery records between 2006-2011 from California"
Not to put a spanner in your works, as I agree your goal is worthy of a lot of effort from the general public. But, how exactly did you get access to this data, and is it legal?
Additionally, from my point of view, even if my medical "records" were de-personalized and made anonymous, I still want to have full control over who/what get's access to it.
Hi zo1, its perfectly legal, in fact the entire program (HCUP) is carried out by the US federal government agency itself (Agency for Healthcare Quality and Research) for last two decades. We have only made the system available internally to a select group of doctors who sign agreements with the US government. Misuse of this data is punishable by a felony. I have personally met the director of the agency and they know about it, and have seen a demo.
Regarding your point second point, the patient ownership of the data is not very well understood legally. Since most of it is transactional in nature. E.g. consider the Sorrell vs IMS Health judgement by the Supreme Court few years ago.
Following paper in Harvard Journal of Law and Technology gives a good overview of the issues.
By suitably modifying queries to make them "differentially private" (technical term), one can allow queries on the data set which have an arbitrarily low probability of releasing personal information.
Building a database which can be locked down to differentially private primitives (differential privacy composes) would allow researchers to partially unlock this data while ensuring that your medical records are private.
As long as we choose epsilon (the privacy parameter) sufficiently low there is then no ethical need to ask a bunch of fickle data points for permission.
We do something similar. We precompute/aggregate exhaustively by following certain aggregation strategies. The aggregated statistics are further processed to ensure privacy.
Differential Privacy cannot be directly applied since the underlying assumptions are too strong. An important consideration is that the error/noise added is independent of the answer. Which means that the system becomes unusable for almost all queries other than general trends.
By restricting the query structure, we no longer need large amount of noise. Privacy of hospitals and providers is also very important and cannot be encoded in the Differential Privacy framework. Again this is still a hotly debated issue. But even the most vocal supporters of differential privacy agree that it might not be directly applicable for healthcare domain.
Following are some of the paper that discuss this:
So I'm not really an expert on privacy (my main interest in differential privacy is avoiding overfitting), but isn't differential privacy by definition necessary on an individual level?
I.e., if you don't have differential privacy, then by definition there is a de-anonymizing query and you can get PII out.
Privacy of hospitals/providers is a separate issue, and yeah, it's pretty clear that differential privacy doesn't work for them. Thanks for the links, I'll check them out.
Edit: after reading your second article, it's deeply misleading. They assume that to compute a mean, one must compute 2 queries - sum(x) and len(x), each of which must be differentially private. But that's totally wrong! You can in fact run the query mean(x) + noise, and this last query itself can be differentially private.
The article also notes that queries on small data sets require more noise to be differentially private, which is totally true, and obvious.
This also, however, ignores the fact that most statistical inference drawn from such queries will be nonsense even without differential privacy. See, e.g., this article for an example of why: https://www.chrisstucchio.com/blog/2015/ab_testing_segments_...
This is a very bad critique of differential privacy.
The second article is undoubtedly flawed in several aspects, that's why I put clifton et. al. first, which I think lays out the case for studying non-formal models.
Regarding
"""This also, however, ignores the fact that most statistical inference drawn from such queries will be nonsense even without differential privacy."""
This is not true. Just because the number returned by a count query on a very large dataset (~100 Million visits) is very small (~100) does not automatically means that the result is nonsense or can be disregarded as error. Doing that requires understanding the query and a hypothesis with good prior on expected outcome. E.g. intersection of two rare diseases. Where you would otherwise expect it to be very small, but there might be an underlying genetic reason / physiological process which might lead to higher prevalence. Or a group of hospitals using tainted batch of medicines leading to unexplained increased mortality.
Consider this paper where there were only 1000 cases (only 248 strokes) per 1.6 Million patients (even larger if you consider the entire 20 Million patients present in the data). However in spite of the small number the authors showed that the increase was statistically significant by comparing with same period a year later.
Again I am not denying what you wrote in the blogpost. But in medicine and the analyses for which such databases are used, the investigators have access to very good priors.
Out of all the proposals I'm reading, this seems to be one of the most realistic, for one reason only: this is something where a little money and a lot of computer science knowledge can go a very long way.
This data is available through Agency for Healthcare Research & Quality HCUP project. Getting access to it is straightforward if you are affiliated with a teaching hospital and/or a university. There might even be some researcher at your institute who already has access to it. Getting access to it as a private entity (E.g. a startup) is more challenging and often requires a stricter review.
Explore game theoretically sound ways to start new companies as "worker co-ops" and to reward participants. The idea being to aim toward a future where there is a large gray area between "employee" and "entrepreneur."
The companies formed don't have to be start-ups per se: they could be as simple (and low risk) as a group of people who build web sites and apps for clients, but would like to have a stake in it rather than just "working for the man."
Since it is research, it is ok if things are explored that would require changes in laws.
I can imagine a site (or better yet, a standard protocol) -- sort of Kickstarter-ish, but different -- where people start projects, gather contributors, "pay" those contributors by giving them equity (or actually pay them, if the company is bringing in revenue).
There would be ways of holding votes (presumably voting power is weighted by how much you have contributed), and various checks and balances to prevent participants from gaming it. Everyone would probably be able to weigh in on the contributions of others, in a carefully designed way that encourages everyone to be productive, cooperative, and creative.
The idea is that there are a huge number of people out there with talent, creativity, and a willingness to take some risks, but most of them aren't really entrepreneurial types per se. This would be both a matchmaking site for getting people with complementary talents and personalities together, as well as a service for handling equity calculations as people contribute hours (and various other things that streamline the process so people with talent can spend their efforts using their actual talents).
Obviously there is some overlap with what Y-Combinator does, but this is a way of scaling it out much larger, in ways that could dramatically increase innovation as well as increasing job satisfaction.
I worked this out in great detail, it took about a decade to get the bugs out. The problem is that it works well for founders and other people interested in the mechanics of running a company but it emphatically does not work for people who would rather 'just be paid'. Not everybody wants to be an active participant in the ownership portion of a company, plenty of people simply want a job and a paycheck and that's it. (And to make it explicit: there is absolutely nothing wrong with that, just like not everybody needs to know or wants to know how to fix their car in order to be able to use one.)
I have no doubt there are people who would rather just be paid, probably the majority of people. But that still leaves a lot of people to do this sort of thing.
Also, as the companies grow a bit larger and mature, they should be fine for the "just want to be paid" folk, as that is basically how it would work.
Also, I would hope that most of the mechanics would just be handled by the system. At least the tedious stuff. Stuff like voting on company/product direction, and rating your peers seems like something everyone would like to participate in, especially if it doesn't become political (in a bad way).
There have been some attempts at this - assembly.com and one other company that I don't remember the name of. Has not worked out yet, as the contributors usually lose focus after a while
I wasn't familiar with assembly.com. I think that it would be a challenging thing to get right, so that such issues don't tend to occur, but it should be doable. One thing that is probably needed is people who join as project managers, to keep things on track and provide reports on each person's commitments, progress etc to other members (all of whom can rate the contributions of one another). When starting out that may not be necessary, but after it starts to gain momentum you may well need some additional help to keep the momentum going.
While reading "The Dispossessed"[1] and thinking about the way syndicates work there, I had this (generic, not-worked-out) thought about some form of protocol for creating co-ops of the kind mentioned, and then running and enforcing the rules (including e.g. equal dividend payment / whatnot).
I wonder if someone has considered implementing some kind of a co-op system in Ethereum (where contracts are transparent and enforced and no single node is responsible for running the code). :)
Hmm, taking ideas from anarcho-syndicalism and implementing them in Ethereum....
Thanks, looks interesting, will check out their whitepaper!
Interesting architecture in regards to their "General Hook Middleware" and other ways of implementing "middleware contracts" (for things such as company payroll). Curious stuff...
edit yeah, this is interesting. Do you know how close this is to a working implementation? I see some "pre-alpha" screenshots etc.
Low cost, high density housing that's safe, clean, and comfortable. I think this goes hand in hand with the basic income research: what good is a basic income if any decent housing is more expensive than the income? (Alternatively: how do we fund basic income when the basic cost of living is so high?)
I think there's a lot of ways housing could be made cheaper via building materials and techniques, architecture, technology, policies, and processes. One of the issues with extremely cheap housing is keeping it safe and desirable for tenants, e.g., dealing with drug abusers and dealers (which frequently make our current low-cost housing dangerous).
I believe that Ikea had a model home that they were building that was a practical minimal shelter. Perhaps there are ways to extend that?
Also, I remember that there were a few companies in China and Australia that were experimenting with 3d printing of concrete homes. Perhaps that could use a little help as well?
It's only positional because density is limited. If the market were allowed to build up wherever they wanted each "position" would basically be an infinite resource...
Of course if one could build anything anywhere, it would be unreasonable to demand the water/el/internet/sewage grid to be extended there at no costs, so some alternative mechanism to ensure such services would be necessary.
Same goes for other communal services like healthcare and education, I guess.
I think this is important for 'average' dwellers and not just low-income folks.
My wife won't live in an apartment ("high-density") area because groceries are hard to get inside. Car-->fridge distance matters a hell of a lot to her.
> Car-->fridge distance matters a hell of a lot to her.
That's interesting, is there a particular reason for this or is this a nice illustration of 'first world problems'?
To have both a car and a fridge, the money to stock the fridge and then to trip over the distance from the car to the fridge (which can easily be made light enough by using a luggage trolley or by asking the s.o. to assist) as a major stumbling block in the selection of housing is something I never considered possible. I see a lot of good reasons for wanting to live in an actual house rather than in an apartment building (I'm living in an apartment building at the moment for the first time in 40 years) but that one never occurred to me.
I'd agree -- as someone living in Europe, the idea of grocery shopping by car itself is already pretty luxurious.
Worrying about the distance between car and fridge sounds silly when most of us are used to walking half to one kilometer (.7 miles) to the nearest supermarket and taking everything back by foot. Several times per week.
Not to mention that many apartment buildings don't have elevators...
> as someone living in Europe, the idea of grocery shopping by car itself is already pretty luxurious
Wait what? As European students, my girlfriend and I already do groceries by car. My parents and her parents as well. My grandparents as well. Unless you happen to live next to the supermarket, everyone does as far as I know. What country are you from?
To me it doesn't sound luxurious to do groceries by car, it sounds like poverty to have to go by foot for a kilometer while carrying heavy grocery bags.
Clean water and/or sewage. Almost a billion people don't have access to either, because the infrastructure doesn't exist for municipal sewage or water treatment. Huge numbers of people die from preventable diseases, simply because their water is fouled.
Until cellphones, most of the world didn't have reliable communications, because nobody could build the necessary infrastructure. Wireless changed that. We need the equivalent breakthrough for water treatment.
Dean Kamen is doing interesting stuff in this area (a stirling-engine based system for water distillation [1]), but his approach is still limited by costs and distribution.
Solving this problem would literally change the world, and unlike many of the suggestions here, it's an area where there's hope that a small research team could make a dent (for example, Kamen wanted to raise $1M for Slingshot.)
There's a ton of money that goes into research of well-known plant species like rice, wheat, maize etc.
However, there are myriads of species where we know not much beyond the taxonomic assignment and getting funding to research these species is a drag ("How do you want to monetize that?"). Yet there's a wealth of novel resistances, food sources, medicines, biochemical pathways etc. hidden away in these species, there's just no funding.
There are some companies which sponsor small research projects into organisms like that (edit: mostly for advertising purposes), for example PacBio awards one genome assembly project of the "most interesting species" (read: underfunded). This species won last time: http://www.nature.com/nature/journal/vaop/ncurrent/full/natu...
It's a weird tiny grass which can completely dry out and then come back to live, imagine what we could do when we'd take this pathway out and put it into wheat! Completely unresearched (Google Scholar has 122 papers mentioning the name, edit: Oryza sativa (rice) has 387,000) yet massive potential.
It may not be the best fit to your funding scheme because research like this goes far beyond 5 years.
Disclaimer: I work in plant bioinformatics (canola)
Later Edit: There's also experiment.com for crowdfunding research, but it's really hard to get a project crowdfunded if it's unpopular (compare: "We have this tiny plant we know nothing about" vs. "We have this tiny plant that may cure cancer", maybe the plant you know nothing about does that too?). YC could fit its funding scheme above "single project" and below "decades long funding"
Scaling and automation of biological, medical, and biomedical research. There are huge tracks of research areas which essentially are technologically backwards. From poor usage of statistics (see all the the replication issues in cancer research, psychology, and etc) to the lack of high quality programming tools to efficiently solve problems. More startups building cheap, applicable, and open-specification hardware with open source software could be a huge boon. The hardware and equipment situation of many scientific labs is stuck in the 90's at best and often based on equipment developed in the 50's and 60's.
Having just finished a master's in material science in applied bio-mechanical research I am personally aware of the lack of basic equipment and tools. Machine learning techniques for example could be applied to many areas of experimental research in biological settings but most labs do not have the time, money, or expertise to apply it. Another possible option would be a sort of "professional software development co-op" for scientific research. Being able to buy (subsidized) development resources might be incredibly helpful. The cost could possibly be offset by forming partnerships to commercialize valuable research, similar to a YC model where the few successes could fund the other.
The system for basic scientific research also seems broken. Much of the data used in many fields (especially nuclear research) is based on (very excellent) research from the 50's. We're essentially riding on accrued capital which will likely reach a limiting point in the coming decades.
Software engineering. There has been very little experimental work done here, simply because it's incredibly expensive to perform interventional research. But if we had solid research which said "software development method X produces software faster and with fewer bugs than software development method Y" it would be a great boost to startups, and YC is ideally positioned to perform such research given that it is surrounded by (and funds) lots of software development projects.
I agree. For how much scientific education many software developers have, it's fascinating how little we actually understand about defect rates, productivity, and a whole host of other issues.
Academic research/funding is broken - it'd be cool if there was a way to create something to fix the incentives.
I think that's kind of the goal of YC research in a narrow sense, but it'd be interesting if there was a way to align the incentives so scientists collaborated more instead of hiding research so they don't get 'scooped'. At this same time this could get rid of the academic journals that charge huge amounts of money for access to publicly funded research by selling prestige to desperate researchers (so they can get more funding).
If research was somehow more like open source development where people got credit for their contributions and hypothesis, experiments and papers were worked on in the open (all open access) maybe science and research would be better.
I think scientists want this and universities would want this - there needs to be a better way to get credibility than publishing papers in famous journals.
Science Exchange (YC S11) is solving this problem. The basic idea is that instead of the "authorship bartering" model that's pervasive in research today we move towards a market-based collaboration model. Basically we encourage researchers to collaborate because it's financially lucrative to do so. By making researchers wealthy they're able to release themselves from being dependent on prestige publications for generating funding. As a result of this we can start to deprecate the somewhat toxic "public or perish" incumbent system. If your entire livelihood relies on being published in a high impact journal you're incentivized to do anything to make that happen. That is not a system that promotes accuracy of information.
This sounds like a good idea and is solving part of the problem (getting funding), I suspect solving prestige will be harder though - it's not just money that drives it (YC may be in a unique position to be able to grant prestige).
Scientists want to be published in 'Nature' because it's good for their career (citations) and respect from peers. If there was a way to fix that incentive where it was more prestigious to do the research in the open and by collaborating (like FOSS development) that would be a big win.
My startup, Thinklab, is trying fix the incentives. The concept is to partner with science funders and help them distribute their money in a way that rewards scientists for openly sharing their work in real-time while collaborating with each other over the Internet. The first simple step is requiring researchers to openly post their proposals in order to get funding. A portion of funding is then set aside to reward valuable contributions from the scientific community. We would love to have the support and participation of YC Research.
I think that's cool and could lead to more collaboration/others also being able to easily reproduce experiments too (assuming the tool for them collaborating over the internet is really good and they want to use it).
The tricky part is that the current system may act as a filter where the best people will still try to hide their methods and get into the existing Journals because it's best for their career - so you'll get lower quality requests for funding.
I don't understand why Universities don't work together to push towards open access, but I guess they're interested in Journal prestige too.
As a side note there's also the issue with papers hiding critical components of their research so they can't be reproduced (so they can make companies later) - this is really not in the spirit of science.
> The tricky part is that the current system may act as a filter where the best people will still try to hide their methods and get into the existing Journals because it's best for their career - so you'll get lower quality requests for funding.
Yes you could be right about that. It's possible that in the short term an individual funder could see a drop in the quality of applicants -- but I think this will be offset by massive increase in overall impact. I think science funders underestimate how much power they have to compel change.
"get rid of the academic journals that charge huge amounts of money for access to publicly funded research by selling prestige to desperate researchers (so they can get more funding)." Quite well said. I think we really need two things to make that happen. 1. provide cool alternative to private companies selling prestige; and 2. exert more pressure by people from outside on the academia administrators and researchers. The current academic system is rotten, but it is, unfortunately, quite stable.
Addiction. It affects so many of us-maybe all of us-greatly. Our legal system, education system, healthcare system are all heavily burdened by it, and they all treat it differently. Good solid science (and a health dose of PR) in the name of ending addiction could go a long way towards making the world a lot cooler place.
Nutrition. A lot of this might just be meta-research, but I feel like we still don't really understand what an optimal diet looks like, and if optimal diet might differ from person to person. We learn more every day about how big of a factor it is in disease processes, especially the two biggest ones in the US - heart disease and cancer.
I think preventive care in general is a great area to dive into, and this seems like a good place to start.
Not only that, it's hard to think of an area with quite so much "common sense" and "bro science", but also an area that should be so easy to research. Take people away on a residential and you can control exactly what they eat, their exercise etc. One of the nice things is that it'd be hard for anyone to patent findings either (I imagine).
If I ever have enough money to dabble in my own research, it'll be there. How great would it be to be able to distribute nutritional advice to the world that was actually true, rather than whatever fad diets are being dreamt up or common-sense "wisdom" is being passed down.
Agree on this. Especially research on Soylent like nutrition products for third world countries. Or the cheapest way to build muscle. Or a Crossfit alternative that's free and does not need equipment.
CrossFit is already free - they've been posting the WOD on their blog for free since forever. You could just take their body weight exercises and pick one randomly every day and you'd be doing CrossFit.
hence the research. Yes - I have been following people like Layne Norton, Lyle Macdonald, etc. for years. And Crossfit has a huge number of critics for its program. It would be see results of research on a safer, better program/diet.
But even people who read research diligently (like Lyle) are only as informed as the research is accurate. Look back to what Lyle was writing 16 years ago (strict caloric arithmetic, the discovery of leptin will make dietary obsessions obsolete, microbiome doesn't matter, etc.)
agreed - which is why I think YC should spend on fitness/nutrition research.
I personally think this has the potential to be a moonshot - if an unbiased entity (with no ties to entrenched interests) would spend money on research.
Along with clean nuclear fusion (already well funded), this is another great idea. I think there are already a couple of big projects working on this...
A little robot that would patrol a beach picking up small plastic bits 5mm - 15mm in size would be amazing. There is so much of that crap on our beaches. It's crazy. Also making a robot move on sand, rocks, pebbles, keep out of the way of people and dogs. Navigate sand and rivers and the tide. It's hard for people to do, we see people fall over, or get wet feet from a big wave washing up the beach or get cut off by the tide quite often.
I have been thinking robotic environmental cleaning a lot over the past years too. Along with raising awareness this is critical to implement in order to get back to a point where existing pollution is negligible.
I want to see someone do forest gardens / permaculture at scale. Right now the largest projects have only been a handful of acres. I'd like to see someone setup a forest garden that's 1000+ acres, so that we can learn more about the economics, viability, and best practices for actively managing entire continents to passively produce food the way the native americans supposedly did.
I realize this is out there, but I also think it's something that's viable from a cost perspective, and also has the ability to capture people's imagination like few other things. I'd also like to see an emphasis on promoting fungi in the ecosystem. Right now there isn't a single park in the entire world that's actively managed to promote the growth of edible mushrooms. And while setting that up would take 20+ years, it wouldn't be that expensive and again I think there would be something kind of magical about it.
Bureaucracy is responsible for a lot of corruption. Getting things done in the government is meeting a lot of people and filling a lot of forms. Bribing is sometimes the shortest path to get work done.
Having interoperable data systems, forms that can be filled online and an easy way for citizens to understand the law and fill the papers without a human touchpoint would make things much more seamless and transparent.
P.S. - Low level corruption might not be such a big deal in the developed countries but is an efficiency drain on about 70% of the global population living beyond that.
The root has more to do with the socio-economic status of the population. In a country where everyone made more or less the same and worked about the same and respected each other the same, corruption would be an anomaly.
Long term solution is to improve the fundamentals - education, inequality, access to basics.
I think you should research urban agriculture. Take Detroit as an example because I grew up there and return often. Detroit currently has hundreds of vacant acres of cheap land that you could leverage to boot a new city. I think there are large opportunities in urban agriculture. Parts of the city are a food desert, until three years ago there wasn’t a single supermarket in a city with close to 700,000 people.
While best practices are pretty settled for regular farming, it’s decidedly not so for urban Ag. Detroit has a quite high rate of unemployment, especially for teenagers. There’s an opportunity to provide those kids with their first job and create value for the community where currently there is none.
There are people who want to do it, but the risk is too high because there are so many unanswered questions. Create a handbook of best practices and I think you could use it just as easily in Los Angeles, Philadelphia or Chicago as well as in Detroit.
I've never understood "urban agriculture". Cities have one of the highest land prices due to their proximity to people/commercial centers, I don't know why turning that into a farm, which takes tons of land, makes sense.
On the other hand, I'm very interested in studying the very long-range evolution of cities (500+ years). It fascinates me that Detroit was kind of this pre-Silicon Valley area with high concentration of a single industry and tons of wealth, and after that's left, what happens now? How does a city reconfigure itself when its tax base drops by, say, 50%? How does it deal with its pension obligations, massive infrastructure bill from all the now-not-needed roads, parks, water pipes, city lights, and buildings? In short, how does a city "scale down" in a way that isn't damaging to its institutions or retirees?
Another vote for Aubrey de Grey's SENS. It means funding indirectly fields like alzheimer research, cancer research or cardiovascular disease research with radical and different approaches, more startup-like. Conventional academic research in these fields is slow-paced like things inside a big corporation, and is also constricted/handicapped by the "Publish or Perish" dogma..
The food production is not keeping up with the boom in population, specially in the developing nations. In India, where I come from, farmers are running away from Farming. (Because it's not profitable enough to even feed a family and pay for school fees). We desperately need ways to increase the food production, without increasing the land or water usage.
A way of doing it can genetic engineering (Supercharged Photosyntesis?), but even this solution looks like a distant reality. May be not enough people are working on it or there is a lack of funding here...
Food is the most basic requirement for survival for any life form. Fund this research... YC will be eternally blessed! :D
Vertically plane production within vertical farms, low-watt high-luminance light-emitting plasma bulbs, full-spectrum plant health monitoring systems, fully automated growth (from seed to crop) and harvesting, highly-efficient HVAC systems for cooling and CO2 regulation, aeroponically grown foodstuffs, and electric self-driving delivery vehicles.
A six-storey, 64m^2 (approx. 1 acre) vertical farm can sustain 1,000 people using ~1,500 kilowatts (assuming certain inevitable leaps in lighting efficiency). Or approximately 1,200 homes. New Jersey has a population density of 1218.1 people per square mile. A $2.7 million geothermal power plant would provide sufficient energy for the farm's annual lighting, to give an idea of costs. That power would not be enough to provide the energy for automation or recharging delivery vehicles.
Plants don't absorb much of the green spectrum. If they could be biohacked to use more of the spectrum, the energy requirements could be cut substantially. (Another option is to dramatically improve either laser light or plasma light in the optimal spectrum for a particular crop.)
Additionally, a lot more research needs to be made in finding out how to grow food aeroponically. Currently, I think it's limited to tomatoes, greens, and possibly some tubers.
To introduce minerals into the aeroponic water supply (Nitrogen, Phosphorus, Potassium, Calcium, Sulfur, Magnesium, Boron, Chlorine, Manganese, Iron, Zinc, Copper, and Molybdenum) requires fish (salmon, catfish, tuna), seawater, crustaceans (crab, lobster), shellfish, fungi, seaweed, and captured elements from an open-loop geothermal plant. Nickel might have to be mined. Ammonia can be coaxed from air, water, and sunlight.
It should be possible to grow, harvest, and deliver food based completely on renewable energies.
Furthermore, certain food crops (cotton [dual-carbon batteries], moss, corn, small citrus fruits) can be used to remake infrastructure components. Like a newt that eats its own tail to grow a new one. Moss and corn waste, I believe, can make carbon fiber or stretched carbon nanotubes. Small citrus fruits, if I recall, can be used as a completely organic (cradle-to-grave) polymer substitute.
Voting is very broken. You only do it once a year, for candidates with 2,4,6-year terms. Even if the voting machines aren't hacked the political machines are: it's expensive to run, and tons of interest groups have to be appeased.
Social media is quickly becoming a powerful parapolitical means for distributing resources, but it's also broken. It's subject to balkanization, waves of outrage and other attention span-related problems, and authentication is hard.
Figuring out how to improve democratic resource management for even small (<10k pop) communities would be an interesting problem.
To me most of these suggestions seem like awesome things but not logical fits for "research". I think the key question is not "what is awesome" but instead "what projects would be a better fit for YC Research than for a startup". For example, areas where an open standard would be great, but right now it seems like companies are bickering and trying to control it.
Bitcoin is one example of this. So is the IoT space, both in APIs and in actual hardware designs.
Another situation is things that are fundamentally valuable yet unprofitable. Like drugs that are out of patent and have multiple uses but are not FDA approved for all those uses. Or figuring out how to make more existing research actually public and shared as widely as possible.
Another situation is areas where the existing research establishment is busted. Reproducibility is a big one here. Try finding articles in Nature or Science that you can't reproduce and then make a big stink about their results being false.
Overall though I think every answer to this question should include an answer to, "why is this a better fit for YC Research than for a new startup".
The ocean. Pollution, cleanup, how we are affecting the oceans, how changes in the ocean affect us, and ultimately harnessing the oceans for greater good, preferably without destroying them in the process. Space gets a lot of attention, but the oceans are rightthere and we are (rather carelessly) only exploiting the low hanging fruit.
Nukes. I know YC has funded a few nuclear power companies, but this seems like an area that still needs a lot of basic research, especially with newer reactor designs. Maybe even some more speculative stuff like fusion.
Developing new types of antibiotics. Nobody is really trying, and iChip style devices seem pretty interesting. Even with that, there's still be a ton of work needed to bring new drugs to market. New tools and techniques for evaluating efficacy and pharmacology? For something more exotic, you could study phages.
sama, my apologies if quoting you is too over the top, but the following comment of yours from last year is just wonderful, and it's a great answer to your question:
"I think figuring out how technology can encourage empathy is one of the
more interesting and important open research problems in the world right
now." -- Sam Altman, July 10 2015 [1]
There is one group at Stanford that's been working on the empathy-tech problem. I don't know any of them, but I've read a few of their papers. I've submitted a few of their papers/projects to HN in the past six months.
Most their papers can be downloaded from the "Publications" page.
If you need more lightweight/mainstream news coverage, CBS, NPR, San
Francisco Chronicle, San Jose Mercury News, and similar have all done
articles on the VHIL work/people. Most of these are available through
the VHIL "News" page.
I submitted a few of their papers to HN, but they all received little or
no discussion.
It's ridiculous that we spend so much money and time dealing with legal issues - both in terms of creating new laws, and in terms of enforcing those laws. Because of the delineation between those who create laws and those who enforce them, we came up with a way of encoding law so as to avoid misinterpretation errors. What if we were to automate the enforcement by programming law into a computer rather than writing down the rules in a book? How would we go about accomplishing that? Would we always struggle because computers lack "empathy"? Is there a middle way that automates 90% of the cases while retaining some human control?
I spent a year at the California Department of Forestry and Fire Protection. The number of statutes and regulations for just CEQA (California Environmental Quality Act) [1] for example fills 425 pages. The department also oversees building codes, liquid pipeline infrastructure, wildlands defense, fire & rescue, etc.
Whenever a new or amended statute is introduced it takes months to evaluate if it replaces or conflicts with existing code. If passed and signed into law it can take a year or more to integrate.
Every year a newly-elected legislator introduces a bill akin to simplifying existing laws. Every year these bills never make it out of committee (i.e., they are shelved) because it would require years of pulling committed resources (SMEs, analysts, support personnel and millions of dollars) out of threadbare operations.
Every department in the State draws up a similar analysis, and the cumulative cost scares the legislature/governor. But it's a good way to introduce the green legislator into how the state government works.
The joke in the halls of the Natural Resources building is that there are too many rules and regulations, but there are not enough resources to correct the situation.
F&FP had no central repository of its own codes. Analysts had to use keyword searches of the public California Codes website!
I think programming evolved from math oriented folks, which is why we find it unnatural to reason about law and other societal constructs using a programming language. I wonder how it would look if it were built for an entirely different purpose. It would be really interesting if Y Combinator invests in that research effort.
No, more like replacing judiciary/lawyers with programmers and software. Imagine that law is encoded as a program that validates the conditions and gives out a judgement. Lawyers from both sides get together to program what happened in an event and the computer processes that and gives out a judgement.
We should look at how we can improve the ROI for education.
Millions across the world, especially in developing countries, drop out of school because they (and/or their guardians)
see no benefit from long-term investment in education. Others who somehow manage to stay in formal institutions are exposed to decontextualized education that they cannot realize their full potential.
There will be many different solutions to it. One of them could be a large-scale, technology-immersed learning system that teaches a broad range of topics to students through a vocation. The vocation could be decided based on the learner's interest and the local resources. For example, in northern Nepal, child walk through perilous snow-covered hills and mountains to recover Yarsagumba ("Himalayan viagra"), a fungus with aphrodisiac and medicinal value. Instead, the kids can be educated progressively in details about different aspects surrounding Yarsagumba - mountain climbing, biological systems, business, marketing (where they could sell the collected Yarsagumba), greenhouse and high-tech farming systems, technology, etc. - without disturbing their Yarsamgumba collecting activity.
This is a simple example. Since a diverse topics are being taught and practiced, learners would not be restricted in the same vocation.
The Riemann Hypothesis! Hear me out, because I think it could serve as a great model for modern math research and collaboration.
One thing that really bothered me during my PhD is that even at the Stanford math department it felt like academia is not set up to encourage work on big questions, nor is it optimal for any objective function I can think of. Is it any wonder that many of the recent breakthroughs (Perelman with Poincare, Zhang with prime gaps) came not from establishment mathematicians, but ones who couldn't or wouldn't play the academia game and ended up almost destitute on the fringes, working away at these problems that they knew inside could be resolved.
Now, why specifically is the Riemann Hypothesis a great place to start? Math problems are like a maze. There are many paths to take, and while you can take satisfaction in exploring one part, you won't know your progress until you see the exit in front of you. In the case of RH the branch factor is absolutely massive, but once you pick a direction, it doesn't take long (relatively) to get to the forefront of what's known. But as it stands it is very hard to get perspective on the problem, or know which ways are good or bad. A mathematician who is not an expert in the field may experiment with various ways of bounding Zeta away from zero, but a domain expert would right away point out positive vertical density results for all a-sets, and bounds on moments of 1/zeta(s). Worse: if you have an idea for a new approach, there's no good way to share it because it will likely be too small for the quantum of academia: the academic paper. Building a collaborative atlas for the Riemann Hypothesis would help solve both of these problems, and could provide a more structured way of working on hard problem than even the polymath projects. It'll be a challenging use-case but at least with 150 years' of work there will be no cold start problem. And I've seen many analytic number theorists who are incredibly smart, hard-working, insightful, great with programming, happy to get their hands dirty, eager to teach others, but neglected by the system and lacking an outlet for their skills. PS: There's still no clear reason to believe RH to be true rather than false.
Is this even possible? Should we think outside the box and see if a more tractable formulation exists? Or should we just accept that approximate numerical solutions are what is possible and concentrate on efficient software and hardware?
Our current model of the internet is some kind of hierarchy, and access to it is controlled by a few selected players. While some celebrate their new hipster web programming language libs, the underlying Internet is dying.
The Internet used to be a place where you can pick a remote address, invoke connect() and have a TCP connection. Now you need a Phd to be able to connect two computers on the internet. If you want an address, so that other computers can connect to you, you have to pay. This money goes to people at the top of the pyramid, for real estate they have invented.
Internet companies recently began "protecting us from bad websites", but this is just the beginning of a larger scale censorship, of governments that tell us what is good for us to read and see. I don't know if this applies to you, but at the place where I live, this has already begun.
In my opinion the main unsolved problem of this domain is how to route packets in unstructured big networks. We really don't know how to do this, and this is one of the main reasons for the current structure of the internet. Solving this problem is a game changer.
If everything goes, please research a viable way to open borders globally. Not only is there tremendous economic potential energy there, but national borders are sadly one of the few openly discriminating policies out there, a major source of inequality and injustice. I would love to see progress towards that end.
The endurance of borders and the nation states they seal is an issue of cultural and political inertia, much more than an issue of finding the optimal economic policy to allow them to open up.
Frankly, and somewhat sadly I might add, I think even if you somehow proved - proved - to a nation's voting population that they would be better off economically (in aggregate) by opening their borders, the perceived cultural and political effects of such a policy would prevent it from ever passing into law through any democratic process. Culture and politics are intertwined in a feedback loop, and it will take many more generations to have grown up in the globalised, internet and cheap-air-travel connected world, where almost every person interacts with, makes friends with, works and does business with people all over the world, for this default to finally go away - only when it does can we ever hope for an open borders policy to even be a remote possibility politically.
With that said, I think aiming directly for an open borders policy might be missing the wood for the trees slightly. Might I instead suggest that we pile research resources and efforts into things like better remote working technology, practices and protocol that might eliminate many of the problems (certainly not all - and mostly economic - but many) that an open border policy should aim to solve in the first place?
In my view the world would be a much better place if we did simply have an open-borders policy throughout much of it, but we can get a very long way towards that goal by allowing people to work from anywhere, for anyone, in a completely effective manner. Not only will this actually solve some of the problems that you talk about a lot more quickly and easily, it should have the effect of flattening out the global distribution of wealth somewhat, such that open borders actually do become more palatable politically, simply because the figurative pressure on the borders of the wealthiest nations would be reduced significantly.
In other words, open borders is not something that can ever happen all at once as a result of a policy change. No matter how much research we do into those policies, they just won't succeed politically as a step change. Rather, it's a positive feedback loop of gradual cultural and political change that has to start somewhere. In many ways of course it's already begun, because of the internet, and I believe the next stage to chase is better, really effective, remote working. I think that's where research effort towards the solving the problems you talk about would be most effectively invested.
How can HN promote capitalism on one hand, and open borders on the other?
Open borders is the classic pre-cursor to a worldwide Tragedy of the Commons. Currently, we have problems keeping public parks clean. Imagine if all of the earth was a shared commons, and no group of people could restrict access to any plot of land. It would be chaos. I can think of no better way to destroy the world than open borders.
Look into tragedy of the commons, and see why this idea will never be feasible due to human nature.
Perhaps you can elaborate in how you see open borders as a case for tragedy of the commons? I do not mean that all plots of lands can be accessed by any individual, but more like any individual can access any jurisdiction of any country. A situation more akin to the EU, but in a global scale.
YCR should fund a longitudinal study of a cohort of say 100 entrants into the tech field, and watch them over the course of 4-5 years. Periodically subjecting them to fMRI scans. Of particular interest would be the relationship of motivation/reward pathways (I'm thinking the striatum) during the transition through burnout. Burnout could be quantitatively measured through techniques such as tracking with body motion trackers, etc.
A follow-up study could attempt to use knowledge gained in this with several treatment strategies.
SamA: If you are interested in something like this I know the right researcher, she has incredible scientific integrity, IMO important for cogsci/socsci research especially of late, although she and I have not been in personal touch for years.
I feel like the world runs on several billion dollars with of insecure code, like GNU/Linux, Apache, OpenSSL, OpenSSH, etc. This code will last decades, and YC startups in particular rely a lot on it.
It's not clear that the current processes for writing code and then patching security holes will bring us into the future. Powerful governments and companies are routinely hacked.
Can we move all this code to a secure future without rewriting it all?
In addition to security stuff, it would also be good to see projects get funded that have the potential to save massive amounts of electricity. PyPy might be a good example of that.
I think the way the code is implemented and managed plays as a big a part as the code itself. Look at how many large-scale breaches are caused by weak wifi passwords and other dumb, correctable things. I made a comment to the same effect, I would like to see rigorous research on what causes breaches and highest-impact mitigation strategies.
+1 for YC Research's basic income (and standard for quality of life)
Meta-Research. +1 for @mietek's Meta-research; creating a distributed, open ecosystem (not a website, but a protocol) for publishing and navigating interdisciplinary research.
Medicine. +1 for Bacteriophages @mietek & open-medicine (e.g. Counter Culture Labs - Open Insulin) https://experiment.com/projects/open-insulin. Additionally, cancer and heart disease research.
A distributed, persistent, interoperable world wide web. (e.g. IPFS)
A sensible distributed DOI and identity system for people.
A meta exploration to determine the world's hard problems and impending catastrophies (e.g. www.metaculus.com)
IPFS has some really great ideas. It could enable huge projects of major public interest, like open publication platforms, robust knowledge sharing and corruption-free news service. News sites are currently engaging in silent edits ordered by minority interests; in an IPFS-kind of web, edits of documents would be transparent since you can't rewrite distributed content without people noticing.
Education:
Vocational retraining effectiveness for work for 45+ people. I am not one of them, but there seems to be more and more left behind with no strategy/known good answer. I personally can't figure out how to teach and elevate people to do good work so it tends to be they end up being a long term on social security/unemployment/draining savings. So the question would be:
- What are the most effective ways to retrain people of that age? I don't know of any techniques to motivate people that are not lifelong learners to acquire skills. Is it a demographic thing? This can be warped into 'what's the best way to make expert users of that category' that leads to gamifaction and other things.
Market study:
- What is the barrier to entry to start businesses as a platform?
I feel there isn't enough innovation in pooling the resources to make starting a business. Yes money can solve a lot of problems, but how do you make it so there is a shared stack of resources of logistics, space, pension/benefits to let businesses of <$1MM gross revenue (basically cottage industries). How do we mitigate the risk of the individual to try? What is the best settings for a particular region to accelerate the progress? What can we remove from the equation to increase the growth of wealth of individuals?
Interoperability:
This is something that I was partially pursuing as a research topic- how can we make the transfer of technical information (CAD drawings, models) easier to convert between. A UNIX system works wonderfully because of how piping works, but for highly complex program there is not a wonderful model to build such pipelining without serious grunge work. A lot of value gets lost plugging these things together, and sometimes this work is done by very nontechnical people. There is a lot of work for this in message processing but less in the context of file processing, but really it's a knowledge representation and transfer problem.
I'm going to be 45 in a few years. Your ageism is a big problem.
The thing is, as artificial general intelligence advances, everyone is going to be left behind. Give it 4-6 decades. Anyone essentially Human 1.0 at that point will be irrelevant.
But a few decades out, you can count on your OWN skills to become outdated, and regardless of whether you are a motivated learner or faster learner, it will quickly become very difficult for anyone to keep up. Even the young.
I am not making an ageism argument, it's a request based on my direct experience in my life. Retraining people in that age bracket I find extremely difficult and I cannot find good resources about teaching and helping them in the most effective ways.
The fact that it is difficult to keep up is the problem I want people to look at.
What needs to be taught is that it is POSSIBLE to automate a problem. Many people (not sure why you're picking on over 45s here; it seems like this is true at all age groups) don't fundamentally understand that computers can be used to perform tedious or error prone tasks.
Ask a non techie how much they spent at Amazon last year, and you'll get some mumbling about mint or their bank offering a search option, ask a UNIX neckbeard and get an answer via awk and col, and ask an overworked, underpaid administrative assistant and you'll get an excel spreadsheet that crudely but effectively provides the answer.
I think we should find a way to take the best content from Math Circles and get that to far more students in a digestible format.
Drilling down even further, if the Question is "what is the _single_ most useful thing that can be done to increase future scientific literacy and accelerate technological progress ?" .. then I think the answer must be to teach multiplication by drawing rectangles.
Its a really good antidote to 'math is magic, a set of rules you apply without knowing why'... its based on physical intuition we already have, and it leads very naturally to deeper more abstract thinking :
You can go quickly from counting out areas of integer sides via grouping, to multiplying fractions and decimals, to expanding (a+b)*c, to (a+b)(c+d), you can introduce 1+3+5+7+...+n = n^2 and 1+2+3+..+n=n(n+1)/2 , you can introduce primes as 'non-rectangle' numbers .. all the while improving facility with basic computations, and developing better math intuition.
Kids who believe they are dumb at math can relate to this kind of physical/visual construction - if you play tennis or basketball or fold paper or saw wood you already have a good intuition of area.
It seems not much math is being taught in school before age 12 .. We don't have to wait or invent anything groundbreaking - we can take concrete steps now, we know what works - ie. we have the vaccine, we just need willpower to get it to everyone.
Research on organizations and possible improvements through software -
An important aspect of organizational building is to structure it in a way that enables free flow of information, while at the same time, enable people to accomplish their productive best. Having a good understanding of what sort of organizational structures enable such goals, and what policies and procedures in such organizations can be automated and improved upon, can bring in significant improvements. One of the big reasons as to why poor countries remain poor is that there aren't enough people to employ and build large organizations. The small number of talented and motivated people are handcuffed because of lack of human resources to power their ambitions. Enabling even those small number of people to accomplish something great will lift everyone else from their dire situations.
I'd be interested in empowering people from third-world countries with little access to opportunity. Having started, 7 years ago, basically from nothing (a $150 laptop). I've grown to a 6 figures (usd) fortune today. That wouldn't be possible without the internet.
There are many areas that third-world citizen can tap into and work remotely: Programming, Community Management, Writing, Design, Translation...
Most of third-world workers do cheap work (less than $10/hour), with unfavourable terms and unstable revenue. This kill the potential for growth and savings.
I think the challenge can be summarized to the following point:
- Helping them get opportunities to start.
- Embracing a growth and continued learning mindset.
- Helping them grow their skills.
- Helping them open doors for more opportunities.
- Helping them setup simple and safe structures to save, invest and grow their capital.
First of all, congrats! Could you talk a little more about your story of going from $150 laptop to six figures? Also do you live in a third world country? I'm curious about your perspective. Thx.
How about carbon sequestration tech? Seems like the world is starting to move towards renewable energy generation, but that alone won't prevent the adverse effects of CO2 levels rising.
A rigorous survey of what the current possible carbon sequestration tech and methods would be really helpful. Another organization might already have this survey done, but I haven't seen anything online (in my admittedly trivial search efforts).
There is a lot of information out there on this topic in both the scientific and patent literature (search on google scholar).
On this topic it would be great if someone could take my idea for solving GHG emissions via some simple financial engineering and flesh out the concept [1]. Glory awaits :)
1. A lot of industries are concentrated and monopolistic, which leads to both issues of high prices, control and lack of local resilience.
So how do we build a more decentralized economy ? What are the missing building blocks ?
2. Our current remote-work isn't good enough from the psychological perspective. No eye-contact, no body language, isolation, etc.
But solving this is one of the only things thing that can bypass politics as a way to decrease housing costs and transportation costs , which are ~50% of our living costs.
3. Is it possible to go to some empty space in the u.s , and establish a city from scratch , a city that would be purposely designed for low living costs ?
4. There was a study , in 2004, measuring happiness, which showed the Amish we're considerably more happy than regular people , and as happy as the "forbes 400" billionaires. How do we create such happy communities, while still living in a technological world ?
Re (1), I agree there are desirable properties of a less-concentrated economy where every industry isn't controlled by 4 companies and the entire Internet doesn't run on AWS, I just don't see that as the future.
If you agree that technology promotes various kinds of efficiencies (e.g. Uber), then companies who can invest the most in their technology platforms will win the market. Net result: when more investment is required for market participation, capital intensity goes up and you get industry concentration.
I think about this a lot as a matter of personal career planning. A lot of my friends work at Apple/Google/facebook and I don't, and I can't help but wonder whether many sectors of the American economy are going to be dominated by megacorps vs. clusters of small (inefficient, but where I prefer working) firms.
Sure i agree, big business will grow. But by just having some market share , decentralization can have a big effect.
For example, ebay, a mostly decentralized business, keeps Amazon in check and somewhat limits it's power and offers an alternative - and it probably forced Amazon to become a platform for others. Same goes for solar panels, Android roms, forums/blogs vs news, etc.
Number 3 doesnt sound like a good idea. It seems like a way to ghettoize a large part of the population. The fact is that there will always be competition to be living physically close to the best community, so a low-cost urban center in the middle of nowhere would turn into a place like Newark.
One of the hardest skills to ever master is the art of putting oneself in others shoes. It's hard to imagine how it feels to live like a woman in a man dominated world because you can never experience that. Our solution to most crimes is to restrict ones freedom and hope that somehow that person will figure out why what he did was wrong. Does it have to take 15-30 years in order for someone to learn from his mistakes? Can we do better?
I believe we can. We should be able to put someone in an entirely different life by using hypnosis, VR, drugs, etc., and let them live a life so s/he can see the other side of world and learn things on his/her own. What sort of technological changes do we need to enable this? How do we model that virtual world to make people believe that its real? How would it impact someones life after such an experience?
research research. there are a lot of things wrong with publishing and accessing research that really shouldn't be the case: public access to research is locked into expensive journals, while we already have tons of cheap distribution methods; published research almost never includes full datasets for third party analysis (storage is cheap, and distribution again); we never see research with unexciting results; researcher tools are crappy and expensive (there needs to be a free, user friendly, cross platform SPSS). The scientific method is great but it could sure use a tooling overhaul on all levels.
Send us mail at contact@satifer.com if you want to talk about any of these. We're working on fixing "we never see research with unexciting results" and a few others.
We've set up an email list, we're hoping to get some conversations started soon at satifer.com
Queensland University of Technology (QUT) has a research project called CAUSEE where they have tracked thousands of Australian startups to find out how many got funded (a very tiny percentage of all startups), what they were planning to build, whether they succeeded or failed, how quickly they got going and so on.
Their justification was that too often we look at successful startups and then do whatever they did. Which isn't always bad advice, but there's the danger that it might be like looking at lottery winners to see what they "did" to make their win. How do we know that the stuff that those startups did actually made a difference?
For example, the CAUSEE study showed that (in Australia) retaining a lawyer is a good predictor that the startup will fail. Accessing any government service other than the R&D tax concession is a good predictor of failure. Writing a step-by-step business plan is a good predictor of failure.
But that's a bit Australian-specific. Is it the same in the USA? Is it the same in Europe? Is it the same in developing nations?
Should YC Research study scientifically what makes a startup successful in different economies? It would tie in nicely to YC's other work!
It would also help guide investment (and government policy) if we had hard data on what works and what doesn't.
I really think YC Research, given its mission of funding long-term fragile research, should think hard about reworking the fundamental institutions that we live in now.
For example, what would a school for children look like if it was re-imagined without the century of cruft that's built up. What do we do with children and how do we judge what we've done.
Another project might be re-imagining a hospital. Due to economics, advances in technology, current scientific advances, etc, we've come to a rather agreed-upon notion of what healing looks like that is only one answer in a huge space of possibilities. What would it look like if people truly, actually healed and maybe weren't worked on solely as the "medical bodies" they are now.
Hillary Cottam's career (link to TED talk, sorry: https://www.ted.com/talks/hilary_cottam_social_services_are_...) points in the direction of the type of thing I'm thinking of. These kinds of services seem fundamental to how humans work (we have to learn, we have to heal), so we should really use this opportunity to understand these things deeply.
The best alternative to antibiotics is more antibiotics - a flood of more antibiotics, more than we could ever use up. The best way to do that is support automation and other efficiency gains and individual projects in mining the bacterial world for novel antibacterial weaponry. That used to be impractical, but now that there is, as of 2015, a way to culture the 99% of bacteria that couldn't be cultured before, it is now very practical to look for everything we might need in biomedicine in the bacterial world. See:
I can't imagine that approaching ecology from a perspective of re-wilding will be of much benefit. The wild is happy to take over a space that isn't kept civilized (for instance, the exclusion zone around Chernobyl), so you quickly end up with a situation where you want to be able to predict the consequences of any actions you take against the existing ecology, which is anyway how we are managing 'wild' spaces.
Unfortunately many of the effects of humans are long-lasting. The elimination of buffalo, for instance, or the loss of most wolf populations in North America which lead to the explosion of deer, IIRC. I am quite interested in the possibilities of selective breeding to replace extinct species, and the major die-off currently happening could have negative effects we have not yet even begun to see.
"how do we suck the carbon from 150 years of burning fossil fuels out of our atmosphere?" Stop destroying forests, stop burning fuels, start growing more plants. The real question is, how to accomplish these.
Transparent lobbying fond that would be used for the interest of majority.
Currently in US there is a problem of buying political power that is hard to counter with standard democracy instruments (e.g. voting). When the specific political problem is discussed (like finance regulation or incarceration rates) there will usually be some corporations or interest groups that will have much higher incentive to lobby for a specific solution than average citizen. Also, they have much more resources than average citizen. This means that in long term there is a trend of political decisions that are not in the best interest of majority of population (you and me).
How to counter this? We can wait until ~50% of population realizes that this is a problem and votes for someone that will solve it. Or we could start a fund that would work relatively transparent and in those cases would lobby for solutions in the best interest of majority. Fond would be mostly financed by community (donors) - people that realized that the only way to tackle this is to group. Most of us have high incentive for making better society and together we have a lot of resources...so let's use it to make our societies better.
This is a great idea. Stuff like the sugar tariff [1] are a perfect example of what you're talking about, a well-organized but tiny minority using the political process to affect an outcome that's bad for people as a whole. I would contribute money to this if it existed.
I think this is the next big thing after basic income (especially when the question of how to pay for basic income comes up).
The land value tax is universally seen by economists as the most fair tax, yet for whatever reason we continue to ignore it in favor of income, sales, and property taxes - all of which distort markets (unlike the land value tax).
The lack of research and real life case studies on this topic is a huge obstacle to this entering the mainstream.
0. Bounties. Bounties for specific small step goals. Even bounties to define these. I like to vote with time, which we often exchange as money.
1. Open source advocacy on why it matters (in the end, it's because human brains will rely on software more and more, and if we cant self-inspect 'it'... bad things). In the short term, OS distros that make the user experience slick. Custom distros that wrap all the best astro/bio/fea are excellent to show others what is possible and get people involved.
2. Things that give power to individuals. All individuals. Even ones we label to say they are not us. Dont rely on adding laws to accomplish it. Closed source self-driving 'connected' cars being an almost exact opposite.
3. Open source hardware. Really, not just hardware we can hack, but hardware we can replicate. (hard problem)
4. (to make #3 happen) Bounties (which others can chip-in to) for 100% open small scale lithography (a goal at a time; 8088@home) and biological computing (pie in the sky, but will happen, maybe faster than is generally assumed; 22k Rat neurons can pilot a F-22 sim).
5. Promote information sources that make the raw data available.
You could research the funding of startups. Or more broadly, how should capital be allocated?
Admittedly I have no idea how something like that could be quantified, and the study of economics has been trying to figure it out for decades. But if YC Research is looking to make maximally impactful investments, it seems like studying investing itself would make sense. There's a high risk of not producing anything useful, but also a huge upside if you get just a little better at figuring out who to fund. YC Research is in a position to bring together people from the worlds of business, machine learning and economics who might be able to make a dent in the problem.
Practically, it seems like navigating the financial investment process is the core skill of YCombinator, so you'd already be pretty far down the learning curve. And you don't have to convince people to adopt your ideas right away, you can try out whatever ideas your research produces simply by making investments. Better capital allocation of all forms could have a really big impact on society at large, if the right systems can be developed.
You are right across the road from the SENS Research Foundation; you should invite Aubrey de Grey and Michael Kope across for tea and ask their opinion.
Creating working, cheap, mass-produced rejuvenation therapies is the best way to save the most lives and eliminate the greatest amount of human suffering.
The most exciting areas right now are senescent cell clearance (and oh look there's a startup called Oisin Biotechnology seed funded by the Methuselah Foundation and SENS Research Foundation), and glucosepane cross-link clearance, which I recently wrote a long blurb on the current state of at Fight Aging! That is at a point at which research is highly parallelizable and even small investments above the current funding will greatly speed the process of finding either a small molecule drug candidate or bacterial enzymes at the Spiegel lab at Yale.
Other areas making real progress and which could be further diversified and sped up are transthyretin amyloid clearance, where the current best approach there is locked up in the glacial GSK development process, and mitochondrial DNA repair, which is another one subject to great gains in speed due to parallelization of research.
Any of these are only a couple of years away from low-cost seed funded startups, and all are and have been dependent on philanthropy to move ahead in the lab. None of these incredibly promising lines of research are in any meaningful way supported by the existing funding establishment, which is crazy when you look at the data in mice and the supporting evidence. E.g. senescent cell clearance even in its first few animal studies has already produced better and more compelling benefits than any of the alleged calorie restriction mimetic or other age-slowing drug development efforts that have consumed billions over the past decade or so.
At least ten of the suggestions that I've read already are better than what I have in mind, but hey, why not...
How do people stay informed in the 21st century? By which I mean informed well. We are all naturally drawn to echo chambers, and I don't think I'm imagining people becoming more polarised over time as they surround themselves with news outlets that tell them what they want to hear.
This will only get worse as more media companies collapse - only the opinionated ones will get the audience to survive... if they even do. There's a business model question but also a broader philosophical question: how can anyone help citizens be informed about the world around them? How is it sustainable?
Healthcare currently consumes 18% of our GDP and is projected to consume 34% by 2040. This will be absolutely devastating for us. The ability to accomplish anything meaningful as a society will diminish as more and more of our wealth is directed towards what is the most wasteful healthcare system in the modern world.
My wife is a therapist and one day we were talking. A patient was discharged from their hospital and I asked who was in charge of follow up. "No one."
Turns out, there is very very little money in "healthcare" while there is a great deal of money in "sickcare".
If we are going to reduce the amount of money in sickcare, we're going to have to start caring for people before they are 60 years old, 325 pounds, have diabetes and hypertension and a McDonalds craving.
Although there will be people not agreeing with this, it is what I do know through my personal experiences and observations, and I will say it: today most of the whole world revolves around fear. Not the entire world but a big part of it.
It is enough to look at the wars still going on, look at the children dying of starvation still on our Earth and you will know we as a whole are far from been with our hearts and minds in the right place.
So what do I do?
I first start with myself, and then when possible help others. I look at myself and I try to understand what I am. My fears and my loves. And I do my best to live in a such way that I do not harm anyone nor myself although sometimes might be hard. There are ways to do it, sometimes not simple ways, but there are ways. I do my best to understand and know that I am a grain of sand in the Universe yet I am a wonder of creation.
Thank you for asking, if I would have your abilities I would invest in programs which help us as a whole to understad the simple truths of fears and loves which has conquered us and leads us from been what we are today: a global nation living on a planet where children are still dying of hunger and wars are killing people.
I can not point to a particular program but that is the direction where we I know we supposed to go first. Not on Mars, not on Moon, not on Venus, or out of our galaxy.
Good luck.
I know it's not easy and not sexy; but as a species we're in the process of transitioning from a centralized hierarchical model (inherently arising due to costs of communication) to a collective decision making model.
Apart from society collapsing wholesale I don't see any way we won't have global consensus building / decision making in a hundred years; and helping speed that along is the most powerful thing you can do for humanity's benefit right now.
OpenHarvest - Research into food growth and optimization
There's a lot of research being done currently, but it's all private research so every new company has to rediscover what somebody has already done. For startups that have great potential in this space, this is a major issue as the it takes a very long time to perform testing of plants(each crop cycle is ~4 weeks) and that burn through their runway easily. Another issue here is that some of the openly available information out there is just for conventional soil farming. There's nothing really out there for hydroponics/aquaponics. So when you're direct seeding new crops in hydroponics/aquaponics, you have to completely guess on the density of the seeds and the amount/frequency of the watering. There's no way to convert recommended seeding directions for soil to hydroponics. There's no more rows, access to water isn't the same since water is easily available, density affects lighting and lighting affect density. This is a huge problem keeping the hydroponic and aquaponic community moving forward with innovation because it's hard to compete with yields when the yields haven't been optimized for your new growing medium yet. It's also hard to test your new growing apparatus or technology when you're struggling to find the most optimal way to grow the plants. Not sure if this is making much sense, kind of rambling right now. If you'd like more information you can find my email in my profile. I've been working in hydroponics/aquaponics my whole career and this is a problem I've seen at every facility.
My wishlist in no specific order. You should consider crowdfunding some of the research- I would be interested in options to support these kind of issues and I'd have no issue with the results being commercialised provided there was some form of "open sourcing" the technology.
* Green energy
* Battery technology
* Water cleansing (ie pollutants from rivers & oceans
* Water purification
* Air purification
* Food production
* Medical technology
I guess most research ideas here will focus on technology. Just a thought: how about supporting research on economics?
YC is in a great position to spawn a small "think tank" focused on technology startups.
A couple ideas: how about studying the effects of software patents on technology startups? What government policies would really support innovation? How should startups approach intellectual property?
Alternatively: how about ideas in industrial organization that relate to startups or Silicon Valley? (I'm sure if you supported research on technology clusters, governments around the world would love to work with you to help nurture startups at home.)
Concrete suggestion: Find an economics professor who you could support as a "visiting fellow" for X months?
Research into future programming languages, particularly into dependent types and future development environments. A relatively small amount of money can have tremendous positive impact on the world in this area.
Sadly there's an economy of scale to performing a Sybil attack on MaidSafe as well as any other autonomous organization based on resource sharing that uses a global reputation system (Safecoin in this case) instead of a localized one. Another knock is using something like IPFS with an incentive layer on top is a significantly more flexible approach to distributed and decentralized storage.
That said, I think there's a huge amount of promise in the field.
Stack Risk. We need to know how much CPU power entities have at their dosposal. If statistics aren't available, find data centers and energy breakdowns backout those costs and use sqftage to find amount of servers.
COmputing power growth rate is likely biggest metric tp assess how the tech industry is doing and possibly better at measuring an economy than GDP.
The risk asdociated with having elephant flow like ownership, e.g 20% controlling 80% could be catastrophic for numbers of reasons:
Lack of fault tolerance of core services
Lack of provacy due to vertical stack ownership
Ability to map users on internet
Ability to break cryptography
Extension of above, ability to control alt currency.
Ability to attack networks.
Etc. we need to get a pulse on this and it is a bigger existential threat than ai in short term.
Stack Risk. We need to know how much CPU power entities have at their dosposal.
That's actually a great idea and almost impossible to make reliable.
We could measure "stack risk" for registered, 100% above board companies with their own data centers and employees, but we can't really measure the risk of some guy in eastern europe with access to a 1 billion node, globally distributed and illegal obtained, Android bot net.
Understanding how the electricity grid works is non-trivial, maybe I am mischaracterizing your suggestion that it is not.
Digital Ocean, a smaller provider, has 64gb memory 6 core intel cpu in a single server[1]. These are wired up into a cluster or clusters and probably plug in to a backbone provider like level3 or a big ISP. Small fish, last time i looked they have a half dozen or so regions with a few zones.
AWS tries to limit data centers to 100k servers, most being between 50-80k servers per data center.with a power capacity of 25-40megawatts[2].
So no, I don't think you could just get a few people to run servers in their house and offset that.
You also connect through an ISP. ISP can turn your connection off. Google is an ISP, runs popular DNS name servers, sells web donains, sells certificates, etc. so, they have a business at every level of the layer from hardware to application.
I guess I just take the positiom that we should know what percentage of power, routing and flow these guys control.
I'm not/wasn't suggesting people running servers in their houses would/could offset data centres. Just as rooftop solar could not replace large power plants.
Your comment made me think that understanding how computing power is employed and distributed, and how that information is propagated, sounds a lot like the distribution and routing of electricity. The problems being non-trivial to understand.
If we consider what the average person uses in terms of computational power, bandwidth and storage, and how many people an individual shares content with, like they would on Facebook, using email, et c., then a server in the home with appropriate software could easily cover that requirement.
If we consider the requirements of publishing content to a mass audience, then individuals running home servers would not cover this (at least not yet). Just as everyone with rooftop solar could not power the requirements of factories.
Sorry I misunderstood, this is an interesting concept. You mean that, we could look at demand and the back that out to get the neccessary total.
This would be super interesting. If you could find, with a reasonable statistical likelihood that demand is D and computing power is C, if there was a large inexplained discrepency you could calculate the size of skunkworks projects.
More and more surfaces are covered with ads, they are nigh on inescapable. In my life I've seen the world go from relatively clean to visually incredibly polluted. I'd love to find a way for companies to be able to reach consumers without having to take over every surface in sight.
YC research brings (a little) money and hackers. You should focus on stuff not being worked on by hackers, or that is un-fundable by others.
Politics and incentives:
The FDA was invented to keep people safe from bad food and drugs. How do you incentivize the FDA to be conservative, but not too conservative? Is there some way that we can allow people to opt-out of the FDA that would be better for everyone?
Mental health:
Psychology research seems to be in the dark ages. It is expensive, ineffective, and often totally invalid (published results are often not reproducible). Can you figure out a new way to do psychology research?
In addition to things with long time horizons, it makes sense to look at things that have little profit-incentive for startups or companies to work on.
One example would be nutritional supplements. It's fairly expensive to do a good nutritional study, and there's no good way to make money off of the results (since companies can't patent supplements). So, no one funds good studies on supplements.
At the same time, finding out that certain supplements have positive or negative effects would be great for humanity as a whole.
This also goes for nutrition in general, though supplements are easier to research.
Young researchers in the medical research sciences. It seems, at least from everything I read, that the NIH et al are heavily incented to keep the inertia and old researchers doing incremental, low-impact studies in charge. Cancer research needs a DARPA, maybe YC can spark it: http://www.nytimes.com/2014/10/03/opinion/young-brilliant-an...
1. Intersection analysis: Get traffic patterns with smart phones. Get traffic accidents and latency statistics by different intersection designs. Determine which intersections are most ripe for improvement. Give free access to departments of transportation. Traffic is a huge problem for society and self-driving cars are realistically a long way off from a total rollout.
2. Programming humans: When you train a neural network, you give it as carefully chosen data labeled as clearly as possible. When you train humans, they listen to a teacher and hope to collect 5-6 useful examples of a concept per hour somewhere in the thousands of words spoken by the teacher. What if you turbo charge that? One related experiment I'm considering is collecting thousands of images of different body language cues and making an app to train my brain (and that of other social handicaps like myself) to read body language more effectively.
Vasalgel, a 100% effective, permanent but (hopefully) fully reversible non-surgical, non-hormonal male contraceptive. It's a variation of the Indian RISUG formulation, being pushed through the US testing gauntlet by a non-profit organisation. They're specifically looking for "social investors" at this time as they approach clinical trials:
> But now that Vasalgel has passed the “proof of concept” phase, it’s time to kick it up a notch — without selling out or losing the social mission of affordability and wide availability. We’re starting an all-out push to hire top medical device people and bring in social investors interested in both moving things forward quickly, and in keeping the end result affordable.
Cheap Virtual Reality behavioral therapy. There is research that show great promise in VR phobia therapy. VR seems to be quite a potent medium for modifying habitual patterns and emotional reactions. And soon it will be a popular hardware.
It would be quite interesting to research the possibility of delivering working (open source? / mental health as an app?) behavioral therapy solutions for home VR ecosystems.
There is a possibility of a great improvement of therapy costs and the quality of life.
It would be great to set up an institute for upholding scientific/research integrity. Have one division focussed on reproducing published results, and another focussed on checking publications for statistical errors, and a third on the look out for academic fraud (fudged numbers, undisclosed funding, citation rings, etc.).
seL4 and similar applications of coq proof assistant. Existing academia is funding some development of tools like coq, and also some core applications like seL4, and more accessible applications will happen as open source. But we need to make these techniques more accessible for larger applications that will affect the real world.
Example: A webserver proved to respond only to correctly authenticated clients (covering the end to end: kernel, server, protocols, for every client).
Also fund making this research more accessible: try, compare, file usability bugs on, and then highlight the different technologies for achieving trust. For example Haskell Liquid types as a lighter weight complement to proving stuff in coq. Maybe fund research meshing these techniques with languages that provide different type-system guarantees, such as Rust.
Education seems like a big one. I know there are lots of big places putting in lots of money (eg Gates Foundation), but it's such a critically important part of every human beings -- well being. And I think there is this deep sense that it isn't working as well it could/should.
YC should fund research into the foundations of the open Internet and web. The Internet was created decades ago as a public research project. Decades later, tremendous amounts of that value was captured by Internet startups.
The thing that distinguishes the Internet and web from other technology is that it is open and interoperable. Its style is just profoundly different.
Individual companies have no incentive to create such technology. They create things in the style of Apple, Facebook, or much of Google's later technology.
The Internet was funded by the military. Unix was created by a telephone monopoly. The web was created by a benevolent researcher. I think it would make sense for YC to fund the next system along these lines.
This kind of innovation takes place on a time scale not accessible to most organizations.
I call it robotworld. It's a giant warehouse filled with robots, 3d printers other machine shop things and a few humans at first. You control the robots via telepresence which gets recorded and can be used for MI later. And you build and ship things. It's a lab that any one in the world can rent time in and work on automated factories. Should be open source open data probably. Maybe it's a marketplace where people contribute old robots or parts and pieces that you rent. Not sure about patent rights if you build something in the lab which is patentable. You rent peoples time in the lab to fix things robots can't video cameras everywhere. If you think it more of a company let me know. I'll be happy to submit it as a startup. Lol.
Honestly, I think something that would be super interesting to YC and Silicon Valley and the world would be creating a technology prediction market. Using markets to gauge consensus about the future is super useful for any company or government making decisions under uncertainty.
You could have contracts about all sorts of technology predictions, such as "Will more than 1% of cars be self-driving by 2025?" or "Will MOSFETs have gate widths <10 nm by 2020?" or "Will virtual reality be a market worth over $1B by 2018?" or whatever you're interested in (and can define precisely).
SciCast.org was a technology prediction market that I participated in for a few years, but it was an academic project and ran out of funding.
Tools for thought, along the lines of Engelbart and Bret Victor and the rest. Especially strategic: helping people think better in groups. This stuff seems to be getting funded some lately, like at CDG, and much of the work could be done as startups, but there's lots of room for more basic research. (I'm biased by wanting to work in this field.)
Gun control. I'm actually largely against gun control, but the issue is we have almost no good research on what factors we can implement to curb violence in our country. What causes people to want to say, do a mass shooting? How do people acquire the guns they use in crimes? (Are new background check laws actually solving a problem, or should we be focused on preventing people from stealing others' guns?) Rather than scattershot regulations to impede gun owners, how can we scientifically target laws to reduce crime?
The CDC is forbidden from conducting research on this, allegedly, so it seems like a ripe opportunity for a private option.
A scientific approach to spreading rational thinking could be world-changing.
If rational thinking became more prevalent, you'd have more allies--more people asking questions like this about how to best direct their efforts toward the right problems. Rational thinking is a force-multiplier. At the same time, problems resulting from irrational thinking would be reduced.
In the early stages, the research could focus on how to teach rational thinking. From there, research could focus on how to pitch a rationality curriculum to school boards, or how to effectively convert adults to more rational ways of thinking.
Small research grants for students/independent researchers.
I'd suggest something like 5k for a 3 months project. Applicant has to hand in a short proposal in the beginning and publish his code/data/results as open source. There are many great research ideas to work on independently today (e. g. in AI), but it's pretty much impossible to get money/recognition for that. For example, such a program would enable students to work on their own idea during the summer break (and in some cases produce something valuable for others), instead of doing some meaningless stuff at an internship.
Water is the most essential resource to our existence (secondary to oxygen), yet >8M people die annually because of water related illnesses. Local filtration/treatment is good, but we need more cost effective and easier to deploy methods to treat water at a mass scale. We're trying to protect the worlds water, but there is a lot more research that needs to be done to solve this problem. Research can include cheap sensors to monitor bacteria in water, cost effective modular treatment facilities, better potable water routing methods and local treatment and disposal of wastewater.
Linux reverse engineering efforts (i.e. Open source drivers to replace closed source drivers). We use Linux all the time, how about supporting its future?
Linux is really important, but it's the best-funded FOSS project already. No other project gets a comparable amount of time by comparably skilled developers.
An open source driver for any particular piece of hardware is only useful for that exact piece of hardware, and maybe a few generations after it. But it's usefulness dies with the usage of the hardware. Contrast with basic research, which stays around forever and can be meaningfully built on for much longer.
Research is more about underlying concepts than about actually building concrete implementations (be they software or whatnot) (though that may be part of the research process - but only as the means to an end).
Interventional radiology (IR). Most people don't know it exists until they need it. IR procedures range from acutely life-saving (strokes, heart attacks, DVTs, etc.) to cancer (burning/freezing tumor lesions). Other procedures have a major impact on quality of life - embolizing the prostatic arteries for BPH treatment, embolizing uterine fibroids, and draining abscesses.
From HN's perspective, this is some of the most technologically advanced medicine. If you've never seen an IR procedure, I encourage you to see an aneurysm coiling. The patient is situated on a table that moves in 8 directions (up, down, left, right, forward, backward, tilt-left, and tilt-right). A C-shaped X-ray machine surrounds the patient to help the physician guide the treatment. The X-ray arm itself is able to move circumferentially to generate 3D images. Note that X-rays have minimal radiation delivery relative to CT scans. The physician will construct a 3D model of the brain's vasculature, find the aneurysm, and stick a reinforcing coil at the out pouching. The entire procedure is performed within a vessel from a single needle prick.
The common analogy is, "It's like a video game." Here's a picture of the setup:
It's definitely far off. But it seems to me this could be an amazing shortcut/hack to achieve a lot of other things here: once we have mind uploading, we've basically conquered death. We can also probably figure out, at that stage, how to run the minds faster or with extra memory or something, increasing the standard human intelligence.
This is also a way to achieve "superintelligence" via human-based intelligence, rather than through artificial intelligence.
That's definitely one of the things the research group would need to research. I have no idea of how to solve the philosophical questions around this, and neither does anyone else yet.
Research into productivity using different programming languages + frameworks.
There is almost no research into which languages are best and a lot of guesswork. This is also directly related to ycombinator itself so you guys have plenty of incentive.
a) Quantum communication has some awesome communication privacy protocols, can make a huge impact on global privacy.
b) Quantum computing and simulation. similar to the startup rigetti that I know was funded, but from a research level. Quantum computation and simulation hardware can be a huge breakthrough for our understanding of physics, being able to perform computations never even close to possible with a classical computer in a decent amount of time.
2) Refreshing education research. How could we, for example, make high quality learning content (think coursera) something that's used by schools around the US, and would not be compromised by bad teachers?
3) Reinventing the current political system. Somehow we live in a time where political figures can honestly say anything and have little accountability --- both of their previous stances on the subject, and future actions taken. we need to do better here. America can't keep choosing political figures cause they are popular. That information needs to be made easily accessible and a rising cultural norm, like wikipedia.
+1 for Meta research. The way we do research needs to be completely overhauled. An outside body could help kickstart this in a way that insiders couldn't. You could also set the bar high and challenge the industry to meet it, whereas something put together by insiders would have a lot of compromise built into it before it ever saw the light of day.
I'd like to see somebody pull out tech at the roots and start over with a new processor or SoC design. There are a ton of problems in the community that all have their roots in the way CPUs process data. There are also a ton of new technologies that become more and more powerful the closer to the metal they're implemented. Even running the whole project as a whiteboard-only, blue-sky effort could produce some conceptual ideas that others would want to follow-up on. I think if this is not done, all the separate techs continue to develop on their own, and we miss out on a huge chance of making a significant leap forward.
Alternative education systems and their effectiveness - which can be applied to 3rd world countries as well.
Self organized autonomy - what if me and my friends buy an idland and declare it a new country, with technology oriented infrastructure, such as driverless cars, automation of government facilities, alternative currency to money? How can we sustain it?
I vote for better building materials with lower total carbon footprint - lighter, renewable, durable, safe, cheap, etc. we need to improve both energy efficiency of homes, cost to build, and time to build. New materials and methods could go a l0ng way toward helping emerging economies and poorer Americans alike build better housing.
> Can racial balance in business, education, military be achieved without policies that promote Affirmative Action?
No, because Affirmative Action is simply a term of art (originating in a set of executive orders mandating such action) positive action taken with the aim of achieving such balance. If you take active steps targeting that balance you are by definition engaging in affirmative action.
Automation of common medical lab tests. Not just for the sick but for the healthy. How cool would it be to every week send a drop of blood by mail and get to see a nice graph on a website how the different hormones and proteins etc in your blood has changed throughout the year. It could help you live a healthier life.
People that advocate and complain-against "hypochondriacs" have probably caused just as much harm as good. I don't have a citation for that, but at least in my case it's made me weary to seek medical attention for small and/or embarrassing ailments/symptoms for fear of being seen/labelled as a hypochondriac. Same goes for psychiatric symptoms. So instead of being safe rather than sorry for going to the doctor, I feel like I have to justify each visit as being serious enough, even when I am paying for it.
Its kindof strange that we see doctors time as so valuable that we shouldn't bother them unless we are sick enough we cant function. Its like taking the car to the mechanic only when the engine has come off because we don't want to bother them with the small stuff. But waiting for small problems to turn big ... Is also a waste of medical resources. And it might be too late to fix them then.
The idea is to go see your doctor when you have symptoms. If you go see your doctor every time some parameter is out of what you consider to be your personal error bands you'd be spending all day in the doctors office. Tests are useful to rule out things you don't have and to close in on the things that you do have once you have something that doesn't function well enough to notice. Until then it is a waste of resources to spend a ton of money and capacity on testing people without symptoms. Eventually those will clog up the system to the point where those that do need medical attention will end up without it.
Seems like a travesty that these incredible classes are created by our best minds, but then aren't continuously offered. (the materials are available for self-study, but not the graded homework, support forums, TAs, certificates of accomplishment etc.)
Coursera, the prof, the university, have no incentive to keep it going. Philanthropists should really fund continuing development of the best-of-the-best MOOCs so students can keep learning, best practices can keep improving.
There is no good central discovery and reputation for MOOCs. There needs to be a place I can go and see that e.g. LAFF is the most popular / highest rated for intro to linear algebra, and here's how much work it represents.
There are universities who start online programs, but what we really need is an online 'degree' grantor who can say, OK, you've taken these online MOOCs from various institutions, I can tell from the fact that you logged in with your finger, your device's camera, metrics like typing style, word frequency for essays, etc. that you actually did the work, these courses are worth X number of points, here's a 'degree' or a few numbers describing the quantity and quality of your work in various disciplines at various places.
And then the best of the best should be able to tap funds to be continuously offered.
It seems more like an object for a non-profit startup (should be self-sustaining but doesn't seem like the sort of thing that should be in IPO candidate or revenue-maximizing, will turn into University of Phoenix)
For example, why is it that many people doubt that humanity is responsible for global warming? Some work has been done on agnotology -- the study of the spread of doubt and ignorance -- already. https://en.wikipedia.org/wiki/Agnotology And then: how can we offset the forces propagating doubt and ignorance? One of the goals of adding annotations to news with tools like https://hypothes.is/ would be to fact-check misinformation.
Energy: The cost of energy is asymptotically approaching zero particularly in solar. While free energy in itself is great, more interesting questions are: what are the limits of what we can make with it? What are the social repercussions of free energy and its corollaries? To what extent can we subsidize yeast factories -- which we know can manufacture everything from methyl halids (gasoline) to LSD -- to create other substances necessary to society?
Geoengineering: Let's assume we can't stop the forces creating the CO2 that is warming the planet in time to avert catastrophic ecological collapse. What can we do to remove that CO2 from the atmosphere? Let's get serious about studying man-made carbon sinks via the iron fertilization of algae blooms:
I know Rick Doblin has proposed a drivers license type system for gaining access to different classes of drugs. It would be nice to see the creation of an evidence-based system.
A lot of the ideas here are good, but each of them on their own could take all the resources of your initiative and you'd still only be able to start solving one issue.
The biggest meta-problem I see is the difficulty organising our thoughts and digital resources. For personal use, I use Devonthink, but it's far from ideal. For collaboration we use a hundred ways like comment threads and wiki's, which are all rather painful.
But I really feel being able to organise our (collective) ideas is the key to solving all the other problems better and faster, so we should work on it first. Possibly the solution looks like a content-addressable, distributed semantic web with privacy and massive federation, on which multiple user-friendly apps can share pieces of data.
Deriving (almost) limitless energy from geothermal wells can enable energy independence for practically every nation. This can be a huge game-changer in geopolitics and concept of nation states. Perhaps on how to do it and then open sourcing the technique.
Got this idea from Manoj Bhargava's project of using graphene cables to conduct heat from the earth and then convert to electricity. (link: http://www.thinkgeoenergy.com/could-a-new-approach-to-therma...) He claims that they have already figured out the basic mechanics for it - i don't know how far it actually is tho.
Applied and computational statistics, especially related to learning/inferring latent representations and causal inference.
Basically, there's a big complicated world out there, and the data we can collect about it using any fixed amount of human labor is really noisy, and so we need vastly better statistical techniques than are usually applied (see: p-hacking/NHST-worship) to help us figure out what the heck is going on.
How about figuring out how to advance Open Education in some meaningful way ?
For example: Would it be possible to create a YC branded online University which confers honorary degrees ? Even if the degrees are not fully recognized would it even matter so long as the professors carry enough weight ? I know this might sound crass but basically it might look better on someone's LinkedIn profile than a "recognized" degree, and perhaps over the long term the University's stats would be impressive since you could be selective.
Perhaps it can be scoped to CS post-grad only as a way to make it less controversial. It's not a very pie in the sky vision either.. just onboard some of the world's top college professors, rotate in unique "star power" professors (examples: Paul Graham, Eric Ries, Brad Feld, David Heinemeier Hansson, Bruce Schneier whoever has the time for a professorship), then build out the curriculum, the web IDE / screen-sharing / webcam platform. If it were me I'd have an emphasis on pair-programming throughout the technical bits because it's an exciting way to learn.
You could think of it almost as another accelerator program but without all the messy bits. This would allow YC to have more of the "full stack". Maybe there could be some kind of a VC-backed scholarship program for it.
You might end up biting the hand that feeds you (Standford, etc) but maybe they wouldn't look at it this way, either way that is how it goes with tech innovation.
Think about all those programmers who applied to YC who were executing brilliantly but who YC dismissed simply because the market conditions were not ripe for their particular startup idea. Now you can capture them in a different part of your funnel and maybe by the time they graduate they'll have a more palatable idea. In my view education has always been YC's core competency, the venture stuff feels more like a monetization layer to me, it's kind of a hack isn't it ? j/k
p.s. All jokes aside, I am merely a programmer, so I don't know if there is actually a market for this, but if it existed I'd probably apply
Virtual prototyping and probabilistic design of mechanical components. This would lower the costs of very big projects like nuclear power plants and make big nuclear more affordable. Small Modular Reactors are going there but there are proliferation issues and therefore the interest in mammoth plants is not dead yet. One tool nowadays is the CAE suite Tosca by Dassault Systems, which can work with both Abaqus and Ansys. May we have an open source Tosca please? One other issue is data homogenisation, that is to say a template to make the most out of past, present and future applied research in the field. They call it manufacturing 4.0 today.
Blockchains are nice and all for automating paperwork, but they're only one component of a much broader set of technologies.
Building the infrastructure of an organization solely through programmed economic incentives makes it possible to seamlessly draw on talent from anywhere in the world without being subject to the limitations of personal bias.
AOs also present an existential risk given that once they're up and running as a distributed network, it can be effectively impossible to hold them legally accountable for anything they do. Imagine someone trying to sue the bitcoin network.
Current work surrounding Ethereum is one place where the field is developing quickly.
Cognitive science and human computer interface research.
1) Every problem you're working on will be improved by better learning systems.
2) The direct challenger to AI's potential is human augmentation. There's an evolving hardware component to that, but the software needs to keep up, and we can greatly benefit from that now.
We're still reading, thinking and learning with computers as if they were books. That has to change. I want to fully realize that potential, I consider it the most valuable contribution I can make in my lifetime. I would jump at the opportunity to fully invest myself in that goal. Please contact me for a deeper conversation.
Better transportation infrastructure to move stuff across the the globe. We currently use ships, planes, and trucks which aren't sustainable from a fuel perspective or good for greenhouse emissions.
A concerted, long-term (~10 year) effort to develop the models, theory, and tools necessary to simulate a prokaryotic cell. This is not the kind of thing that’s being funded by the NIH, given their translational shift, and it’s not the kind of thing that can be done by industry because of the time horizon. Nonetheless it’s foundational to building a quantitatively predictive biology. It’s the sort of thing that can enable entirely new applied sciences and industries. And I think it’s possible.
There’s a certain way to doing this which I believe is key to success. It can’t be about doing molecular dynamics from atoms on up, because of computational considerations, and because that’s useless anyway (one wouldn’t learn anything beyond the known physics and maybe better force fields and sampling techniques). It also can’t be like the kludgy attempts that have been made so far (e.g. Markus Covert’s paper in Cell from a few years back—it’s perfectly alright work, but won’t get us closer to solving this particular problem.) Instead, I think the key is to develop a new layer of abstraction for describing biological phenomena, much like say organic chemistry is a genuinely new abstraction built on top of physical chemistry (which is itself built on top of QM), that hides away the irrelevant details (e.g. most of QM) while capturing what’s salient about the biological phenomena of interest. It would require advances in the aforementioned areas of models, theory, and tools, that are made in concert with the aim of simulating a cell.
Globally open borders. Allowing people move freely (globally), following opportunity, seems like the best way to reduce global inequality. We allow everything else to move quite freely but if you've ever dated a foreigner or god forbid had to flee home you will understand the struggles people face when trying to move across restricted borders. It seems like a study into how globally open borders would affect inequality and the general well-being of global society would be a valuable endeavour.
What are the positive effects that you expect? Saying "reducing global inequality", do you mean something like communism, where everybody shall be equal? Please, don't take my question negatively. I have myself lived in a communist country and I think it had both good and bad parts. Essentially it boils down to whether everybody becomes equally miserable or equally happy.
Another problem that I could see with globally open borders, is that people may not actually want it. They tend to side with similar ones (race, language, religion,sexual orientation, income, mentality etc). This has been shown scientifically in 1971 in Thomas Schelling's segregation model. We are actually witnessing this on a very large scale all over the western world. Have a look at Germany for example. There are about 2-3 generations (~3 millions) of Turks that have not dissolved in the German culture and are actually quite compact from a social networking perspective. People have been like this for centuries and I don't think it is realistic or even fair to expect they will change (e.g. Kurds, Gipsies ...). Even in the US, which has a good reputation of being able to dissolve immigrants into the US culture, we see Chinese and black neighbourhoods.
That said, I do think that it makes sense to do research on how such segregated groups may live together (in close proximity) peacefully and sustainably. I guess that to some extent these problems are a subject of the military science, although it is probably trying to solve the opposite problems as well (i.e. divide and conquer).
Global open borders would create chaos. Imagine if homes didn't have fences or doors and anyone can just walk in & out of your home.
The argument of open borders as "the best way to reduce global inequality" is pure myth. How do you explain all the poor black people in Chicago or all the poor Mexicans in Los Angeles. These are people whose families have been living in America for generations yet are still poor. How about the fact that upper-class Indians are the ones who get work or student visas but the lower caste Indians don't have the equal opportunity.
I would ask people to address the inequalities in their home countries first before blaming other countries. I think the answer is to first address the reasons why 2nd or 3rd world countries remain as they are, preventing the citizens of that country (and the country in general) to become more prosperous. There are so many countries that are blessed with so much natural resources and human talent yet remain 2nd or 3rd world.
I think the reason for this are: corruption, graft, cronyism, elitism, colonialism, etc. Fix this so people don't have to migrate/flee to the few first world countries - which is not sustainable.
My attention was just drawn to this. Yes Sam - our address is 110 Pioneer Way Suite J, phone 650-938-6100. We might be able to do better than Reason's suggestion of tea :-)
I hope there will also be room for researchers to pitch their own ideas. This more focused approach that you've been pursuing certainly has it's advantages, but I'm guessing that the researchers themselves will also have ideas that you (and HN) won't have thought of before.
(I know you're just starting to build out YCR, so I won't be surprised if it takes a while to get to the stage where you can do this. But still, I wanted to put the idea out there.)
Can't help but promote my own beliefs and ideas here a bit. I have started integrating some sustainability concepts and technologies together into something I call a tiny village. I am hoping to spread the ideas and maybe have people use many of these ideas in real developments someday. It is all open source (work in progress) http://runvnc.github.io/tinyvillage
- automated psychology for families (if every home has audio and video monitoring of every argument and discussion we will soon be able to tell what is normal and what is best practise)
- mesh wifi and local networking
- something around either making it really easy for people to move country or work remotely. When companies and countries have to compete for global talent, many things will change from Syria to salary negotiations. I cannot see how to do this but ...
I spent a while reading all the ideas posted here, and I think most of them are not great. Not that they are bad, just they seem like they would be really difficult to make progress on, even with tons of funding.
I tried to come up with some ideas myself, but I also couldn't. Changing the world is hard! Why aren't there more low hanging fruit?
There is one idea that sticks in the back of my mind, and I keep seeing applications for it everywhere. An idea that I think would make the world a lot more efficient.
I really think that there are applications all over the place for automated optimization. This is a common, recurring problem, that there are basically no tools for. And I believe a simple tool could be made that even an idiot could use for their weird problem.
Imagine you are have a bunch of parameters that need optimized. Like for a machine learning model. You could plug those parameters into this tool, and it could suggest the optimal next test to perform. Or if you are designing an airplane wing, or a chemical mixture, or a food recipe, etc. A ton of things have parameters that can be optimized, but there are no simple tools for doing it.
Of course that doesn't fit into the YC research model. It's more of a startup idea, or even better an open source tool.
A way to teach people how to do research (without doing a multi year "apprenticeship" at a few limited locations). Research skills as one of the "R" taught as a fundamental skill.
Like you can learn programming without really apprenticing yourself to a specific master programmer, people should be able to learn do research without doing a PhD. If this problem is cracked, all kinds of non-traditional people will be able to contribute to progress.
Figure out a way to break translational research out of the academic science model at scale! There is currently a huge decoupling of supply and demand in academia, especially in life sciences which is my home domain. Incredibly talented individuals are forced to invest 6-14 years building the credentials necessary to apply for grant funding / self-sufficiency, all the while (at best) earning what a new grad would make if they became a tech in pharma straight out of college. Most people cycle out before hitting their prime, at the expense of our economy.
This decoupling between supply and demand has also motivated other negative trends, such as changes in employment incentives for labs (large push for postdocs / foreign students at the expense of grad students); increasing importance of publishing in high impact, for profit journals (more paywalling of knowledge); sabotage of the peer review process; and a deliberate siloing of what should be public knowledge to prevent "competition" (among other things).
Ultimately I see most basic science being performed in an institute model existing in parallel to academia. In such a model, investigators should own their own ip, would have their salaries uncapped, and would be able to employ qualified people at a fair market wage. In my eyes, the biggest barrier to enacting such a model is startup capital. Most universities give new faculty members a "startup package"; essentially an angel investment to allow them to buy equipment / reagents and pay salaries before they need to support themselves off of grant funding. A new institute would not have the capital required to support this (unless it received extensive funding from foundations / industry / NIH etc).
I envision crowd funded startup packages with caveats. One potential model could mimic angel investment / old school patronage: an investigator receives $X in exchange for a percentage ownership of whatever IP they create within Y years, distributed among the investors. Another model could mimic kickstarter: people crowd fund an investigator with small investments and are rewarded by being part of the process. They are given regular updates, walked through experiments / data, invited to the lab if they are local to see work being done etc. I'm sure other funding mechanisms exist that I have not considered. I would love to chat further about this issue if you are interested; please feel free to contact me.
As another exploration of wealth and inequality, you might consider helping to fund Bill Black's 'Bank Whistleblowers’ Group' [0].
When Milton Friedman argued that 'The Social Responsibility of Business is to Increase its Profits' [1], he presumed that his argument was true in an environment where a firm 'stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud.'
I think it's plausible that there is room for funding referees of the capital game who are not (or at least, less) biased in their analysis and reporting than what is available in our present environment.
Genetic rescue & de-extinction. The Genetic Rescue Foundation https://www.geneticrescue.science/ and Revive & Restore http://longnow.org/revive/ have research projects that are advancing the scientific techniques required to do this. Having YC Research also take an interest in this would help to expand a field that already has the attention of some pretty talented individuals e.g. George Church.
Human beings are now the curators of planet earth. This is the Anthropocene era. We can either continue to intentionally and unintentionally destroy the earth's biodiversity; or through technology we can develop the means to preserve it.
We're not preserving biodiversity because panda's are cute and starving polar bear pictures make us sad. We're doing it because biodiversity is useful to humans and it takes a very long time to get it back when it's gone.
The biggest opportunity for impact would be to find some politically incorrect / heterodox studies in areas where existing power structures have no incentive to fund. Nutrition comes to mind but it's a pretty crowded field. Alternately I think funding more open source legal docs like the SAFE could offer huge social ROI. (Perhaps in non startup legal areas.)
I came here to write "basic income" and was pleasantly surprised to find it already on the agenda.
Other suggestions:
- Desalination / easy clean water
- Related : Cleaning the oceans. This may sound noncommercial, but if the profit-motive needs to be addressed : a worldwide clean-the-oceans tax will become palatable if the situation with the world's oceans worsens.
Remote work. We need research to give us answers about productivity tradeoffs and also specifics like what kinds of jobs are well-suited for remote work? how does it affect the life satisfaction of workers? Does it increase employee retention? What effects do cultural and language differences of employees have on productivity when working remotely? etc...
Hearing aids that sync seamlessly with Bluetooth. It can really transform the way hard-of-hearing people converse on phones.
Hearing aids whose battery life is supplemented by body heat (Battery life is currently ~1 week, and they represent the primary reason hearing aids are so big, even as small as they are).
Smarter or more customizable software to adjust hearing aid ranges and frequencies. Right now I have to go into an audiologist, but it'd be nice if I could see my own hearing data and make minor tweaks without having to pay time and $$ to see an audiologist.
Initiatives to teach mobile app makers about how to help their apps appeal to people with hearing aids. (Hint: subtitle everything)
In short, think of Google Glass but for hearing instead of sight -- and think of it was a product you can sell to the normal hearing public as well. Augment regular hearing, as well as assist with hearing loss. Imagine being able to go to a restaurant and silencing everything but the people at your table. How nice would that be?
Not sure if these are within the scope of what you guys do (a couple are 'out there') but here's some:
- digital currency that doesn't need network connectivity nor blockchain to do a transaction (just device to device)
- systems that actively monitor for suspicious/anomalous network activity and respond in real-time (security that can prevent Sony & OPM hacks) or simply put: security products that will stop the Chinese from hacking US networks
- urban farming
- fish/seafood farming
- batteries from garbage
- products that save rain water - from your home's rain gutter or from the roadside curb - and into a cistern
- a way to lift third-world countries out of their situation so there are no more third-world countries (all of Bill Gates' and Oprah's money have barely made a dent)
- underwater or ocean based communities
- cleaning up the 'Great Pacific Garbage Patch'
- more women/female programmers/engineers (it IS the pipeline and perhaps even the culture)
- an actual, real hoverboard
Are you really going to read all these comments or did I just waste my time writing this??
A Warpdrive. (Just because I am currently playing with GR as a side project.)
Actually serious, I believe the transmission of information between the frontiers of science and the public are broken. On one hand the university system is geared towards producing specialists, on the other hand the media is geared towards spectacular as orthogonal of important, due to a mixture of science journalists being the product of the university system and market forces. So one would need a reliable process to produce the one book to read about X.
In first approximation one could try to produce a four volume series "Frontiers in Physics", were the volumes are "Particle physics," "Atoms & Nuclei," "Solid state physics" and "Astronomy and Geophysics" and then distill it into a single volume. (As a illustration of Universities producing specialists, I could not name the four titles for chemistry, let alone linguistics.)
Genetics and Epigenetics research at scale on humans. Just getting the database together in a machine readable form would be a huge accomplishment
There is a huge (missing) sample problem in genetics. This creates problems in epigenetic research, even with gene variations where there are accurate sample sizes, because it becomes extremely difficult to figure out if it is the gene that it is the trigger or an epigentic factor turning on a gene, and if so, what sort of epigenetic factors are we looking at in the first place. Turning off genes through behavior changes is mostly off the plate without getting accurate data first.
Part of the reason: It is still expensive to decode every genome in a human (around of $250k per person right now) That cost will drop radically in the next 10 years because it is only 1-5 parts that really need to drop in cost to make it around $250/person. Then matching that data to health and other related records en mass is huge- especially because most basic health records in EHR are geared towards billing procedures and short term case discussion, not tracking people, families, and communities.
Furthermore, it is a true ethics issue. To donate part of your full genome to science means parts of your genome will be bought and sold for research purposes. The going rate for simple stuff is about 1k/genome, with more money to the market with more interesting medical histories and family data attached. Creating a really big database that is open means this data stops being in a market and could push for innovation in the field.
I'd like to remind you that we have no idea why the county that Ycombinator sits in is among the top counties in the country for Breast Cancer rates. Just getting an accurate sample size of Marin, San Mateo, and San Fransicio counties's genetics and true medical history could make a huge impact on why people get cancer period, because we have no idea why there is a cluster there.
I'm biased because of my work, but neural interfaces for general purpose tasking or entertainment. Alot of fMRI/EEG/MEG research is focused on medical applications, but the funding cycle is tied heavily into medical related grants, and current applications meditation tasks that are popular generally wont get people excited (I've ran a lot of novice/experienced mediators who have started to fall asleep or become disengaged from a research task and who need continues human engagement to keep them in the mood…), nor convince people that what they are buying is worthwhile, unlike say buying $500+ gaming system and games people play for hours on end without having to convince someone of it working when they just get the immediate benefit. This could also help get more people from software or other engineering backgrounds more exposed to the field now and help push things forward faster.
When people live in peace, when countries live in peace, creativity of people will boost, number one priority of each person would be how to impact peoples live in better way.
Every single research will be shared across countries to make lives of poor even better than ever.
Imagine a world, where you can go anywhere without fear and help anyone, even more, get help from anyone.
YC Research into LSD and Psilocybin would be great. These drugs have been making major impact in the psychotherapy community, especially war veterans and rape victims with PTSD.
Attempt a blank-slate reboot of macroeconomics. Using data, build a framework for evaluating macroeconomic models. Develop and test models without prejudice.
Human civilization operates at a higher power density than the core of the sun. We should understand it. But economics is stuck a natural philosophy, and most of it is actually anti-knowledge.
Success here will unfortunately require significant effort (10-15%) be spent on evangelism, including criticism of existing economics.
The main consumer of economics is government and there are strong incentives to exaggerate the effectiveness (either positive or negative) of economic theories and policies based on them.
You will have to work very hard to avoid politicization.
Recent work in thermodynamics (Tim Garrett, Jeremy England) may be helpful on the theory side; machine learning on the empirical side. To get financial data, homomorphic encryption: https://numer.ai
Focus on things that can't be funded in traditional academia, yet which are public goods and so cannot be funded by startups.
Project that have one or more of the following characteristics:
1) requires a large amount of implementation work and real-world trials before you know if it's a good idea
2) a long project, but if it fails, there may be no publishable intermediate results (in academia, we stay away from long projects unless they are publishable no matter what the result turns out to be; b/c at the end of grad school, or your postdoc, or your tenure clock, you need a publication in order to remain in academia)
3) reform of academic funding (it's not advisable to pitch an NSF government official to fund a project that only makes sense if NSF government officials are unnecessary)
4) success criterion is a matter of taste, rather than objective (or, maybe the true success criterion is objective but too expensive to measure, so taste must be substituted as a proxy success criterion)
5) creation of useful artifacts, rather than new knowledge
6) requires ongoing maintenance in order to remain useful to society
7) more a matter of putting together or implementing existing ideas in a wholistic way than discovery of novel ideas
Examples:
* alternative systems to allocate research funding (satisfies #1,3,4)
* creation of new programming languages (1, 2, 4, 5, 6, 7)
* open-source software development (5, 6)
* creation of new software-supported governance procedures (1, 4, 7)
* creation of new legal structures (for remote working open co-ops, etc) (1, 4, 5)
* direct attacks on the Millennium problems in math (1)
* upgrades to the academic peer review process (1, 4, 6)
* alternative systems for hiring professors (1, 3, 4)
Knowledge stewardship. The world's knowledge is growing exponentially, much faster than the tools to aggregate, catalogue, and filter that knowledge.
Khan Academy is a good start. And Google is, well, Google: you can find what you need, if you ask the right question. We need the 21st century library-teacher-guided learning system.
Would modest funding for legal defence make a real difference for poor-to-middle-class people?
One of our nastier social problems is that the legal system assumes people have realistic means of defending themselves when prosecutors charge them with crimes. But for people without savings -- which is a lot of people -- that's not really true, so they end up taking all kinds of crappy plea-bargains rather than actually contesting the government's accusations in court.
There's legal aid, of course, but it tends to be funded at very low levels, and done by either wildly overworked staffers or contractors paid at such low levels they rarely deliver a spirited defence.
If the down-and-out were given enough money for a real defence, would the law deliver results that are more just? By what metric? And how much money would it take to make a real difference?
Racial equality, especially in tech. Develop a disruptive and innovative way to help bring more diversity into the technology field. How to break down racial barriers to more people of color, especially blacks to enter STEM fields and level the playing field for access to education needed for that.
There are still millions of UK people - young and old - who either do not use computers at all or are barely beyond beginner stage, and therefore cannot make intelligent, relevant use of it. See statistics at http://www.ons.gov.uk/ons/rel/rdit2/internet-access-quarterl...
This is despite the work of Martha Lane Fox's Digital by Default project, very generously government funded and rewarded by a seat in the Lords. The obvious solution was not acted upon, and I would like to put it to you, having already done small scale research on it.
Anita Pincas a.pincas@ioe.ac.uk Univ. College London
Better way of sharing the state of the world (upgraded news sites and/or journals) -
If you open newyork times or any news website, its clear that they are stuck in the 19th century. They are organized on the basis of time rather than on topics/people/metrics. What if we have a news website that's actually a dashboard? It can
- Highlight the current metrics of the world - crime/stocks/happiness/weather, etc.,
- Highlight the currently active topics and people - what's happening/what did people say, etc.,
- Journal the world in different axis - geographical, temporal, personal, emotional, national, topical, cultural, etc., And let people see the trends in each of these dimensions.
How does such a thing look like? How does journalism evolve in that kind of a world?
I perceive that US Citizens feel disenfranchised when it comes to national politics. Is the popular apathy related to our 2-party system or other factors? I believe we have the technology today to effect meaningful change, but it hasn't been done yet, and I don't know why.
- Having had a number of friends doing research in the field of anthropology, I recently got to think it represents a bag of knowledge we could leverage on to answer very important questions, for example, "what is really the role of religion in our societies?". Governments are spending huge amount of money to defend from/attack terrorists that let themselves explode in name of a religion, but no money is spent to understand deeply how this is possible, and whether it is really what is the real role of religion in that.
- "Nutritionism", or whatever other word you like to mean "how what we eat impacts our well-being". We need to eat, but information on how we should do it to be healthier seems very contradictory.
Tamid is a student run non-profit that helps students get internships in Israel. It's different from a program like Birthright in that it facilitates a cultural exchange through business rather than religion.
Another commenter brought up that teaching empathy has been something you, sama, have personally thought is an interesting and difficult problem, and a Tamid-like program is one approach. There's nothing quite like living with a foreign family and working for a foreign business that makes you fully appreciate the perspective of that country.
Although Tamid is Israel specific, one could imagine similar programs for India, Russia, China, etc.
There are a lot of Non Profits who are doing amazing work but are often faced with lack of funds. There are also a lot of people who did like to help but refrain from doing so for lack of motivation, empathy or trust on the organizations. Sometimes people feel a sudden urgency of doing good but because its not as quick as calling an uber, they end up not doing it. That lost dollar could have fed a kid somewhere.
I believe human race can be uplifted with numerous enablers and doers on the ground. But there needs to be work/research done on the best framework to make the whole process as seamless and rewarding as possible.
Companies like google have attempted this with OneToday App. More companies and startups should come together on this problem including YC.
The effect of education.
The effect of proper diet.
The effect of someone following their own path in the field of Mathmatics, Medicine, or Science on society as a whole.
How long it truly takes to learn a language when it is your singular focus as an adult.
How to provide tech seed funding for remote and underserved communities here and abroad.
This is looking beyond what YC Fellowship provides. This is not necessarily looking for unicorns, but what does it take to scale out tech seed funding to remote and underserved areas.
Open source operating systems for the Bangladeshi farmer's smart phone smartphone: a computer used until it breaks and improves a person's quality of life.
What infrastructure is needed to make making open source projects easy for ordinary people to use easy?
+ What sort of tools make developing software for embedded devices that is updatable [securely when and where needed] easier than making software that can't?
+ What does an easy to use security model for a world of updatable embedded computing devices look like? How does it play out in situations with high communication latency and network partitions?
+ How can general purpose computing be integrated into devices and be made as easy to use as a smartphone camera?
I'm wondering if you might think about pilots or creating models. In particular, I'm thinking about how design thinking can create more opportunities for communities to experience deliberative democracy. That's the work that I do, and I'm about to start a project in NYC that focuses on creating an admissions policy that will result in more diverse middle schools. It involves research, but it also puts forth, or we hope it will, a model that other districts can tweak and try out. Here's an example
https://www.youtube.com/watch?v=XIRkVZDILzM
A research on how to unwind grudges after long term conflicts. Specially for large scale conflicts (as opposed to personal conflicts). A notorious example would be the Arab, Iraeli conflict. That is a conflict that affects me personally, and i ve always been worried about how, even if official peace treaties are signed, will the people get over the hate, fear of the other, the feeling of not having been vidicated. And the effect of such feeling on the peace.
A case that has always fascinated me is the german european relations post ww2, same for nippon-us relations, how did these pull it off?
Finding out the hows, and whys, can greatly help in improving global well being.
Seriously. The Space Elevator guys are working with micro-fibers to come up with some kind of rope/scaffording that would support its own weight and that of an elevator all the way out to GSO, but it sounds like there ought to be some early victories where you could take microfibers and create super-light rope and cabling.
I have no idea what the impact would be but it sounds like something very interesting and somewhat useful that the commercial sector wouldn't develop on its own. It could also lead to a bunch of other stuff that'd be neat, like new textiles. If nothing else, we'd learn stuff about what was feasible and what wasn't.
(1) A detailed study of cost in the (building) construction industry.
(a) What are the components of cost? What is their relative size? Labor, land, materials, permitting, architecture/engineering, regulatory compliance,
(b) The housing value chain. Parties involved, who's adding the most value/earning the highest markups. How the cost of commodities (wood, steel) compares to the final, built value of a structure.
(c) Why it costs more to build in some places than others (materials availability? labor)?
(d) The true cost and benefits of various types of regulations, e.g. SF's Affordable Housing Bonus Density Program, various types of zoning ordinances, height limits. Quantifying the cost of regulatory uncertainty in extending project lifetimes. The extent to which a regulatory/uncertainty premium influences investment behavior.
(2) Effects of technology on mental health. I've been happier since giving up facebook and I wonder the extent to which this is a placebo, or something more widespread and generalizable.
(3) Computer security. A study of what actually causes breaches, data theft/loss, and other high-impact disasters/outages and how we can fix it, with detailed cost metrics. Things like "employee malfeasance", "unpatched software", "configuration error", etc. As a sysadmin, I'd like to know "what are the 5 top causes of breaches" and "what is the highest ROI investment I can make in software/procedure/etc. to keep my large-scale computer system safe/confidential" (I wanted to do this one in grad school but never got around to it)
(4) A true study of the early-stage ecosystem. Kind of meta, but I'm tired of people making vague hand-wavey generalizations like "99% of companies fail". How much return do venture funds actually earn? What are their biggest economic drivers? How do VC returns as a whole correlate with other benchmarks like the S&P 500?
If anyone wants to discuss these further my email is in my sig. I used to be a researcher and think of questions like these all the time when walking the streets of SF ;)
The projects being funded by YC (esp basic income) are substantially different and difficult to scale than those of traditional startups, YC or otherwise.
In the interest of reducing noise, are moonshots what you are looking for, or is it anything-goes?
+ Real property. Nearly all research is driven by commercial concerns. Detroit and New Orleans show how brittle that approach is statistically for events we can anticipate because of the scale at which urbanization is occurring.
+ Ubiquitous production of computational artifacts. The analogy I started with was photography: that computing is in an age when heavy equipment, toxic chemicals, glass plates, and holding still for a long time are barriers. But I've since pushed the analogy further back: maybe to oil portraits, or paintings on tomb walls, or perhaps Cromagnon caves by firelight. Now the world is flooded with images.
Crime investigation and forensic tools. Move "CSI" from science fiction closer to reality. Apply solid scientific principles and automation to cut costs and prevent human errors.
For example currently police don't have the resources to really investigate many crimes. So let's build small robots that automatically gather physical evidence from crime scenes such as DNA, blood, hair, fingerprints, fibers, etc. Make them cheap and reliable enough for every small town to afford. Then build automated analysis machines to look for evidence matches in databases and output statically valid match confidence scores.
What we'll do with all the electronics waste we're producing now. Disassemble it? Take it apart with robotics? Is it possible to reuse SMD-soldered components in new things?
This problem is only going to get worse with time.
How about checking into the use of optogenetics to build a universal, minimally invasive neural interface? We have optogenetic treatments that make nerve cells responsive to and emissive of light. Can we treat nerve cells to transmit and receive light over a fiber optics, letting us bridge gaps in existing neural paths? something like that night go a long way to healing all sorts of neural degenerative disease, and possibly deal with paralysis as well.
Perhaps eventually something along the lines of plug and play hardware for humans, if you can come up with some sort of universal spec for it.
At the risk of being too ambitious, I would absolutely love to see YCR fund research on terraforming Mars. Based on the recent climate change data here Earth[1], this is important for 2 reasons:
1) We might be in a position of needing a new home sooner than expected
2) Humans seem to be better at affecting global environments than we originally thought
It can only help us to have a better understanding of what a realistic budget and timeline for terraforming would be.
Resources allocated to interplanetary redundancy enhance the survival of the people "over there" at the cost of you being more likely to die "here." I prefer to amplify the chance I and my loved ones survive, instead of splitting 4's at the blackjack table. Also, you shouldn't be sitting at the blackjack table.
It includes a link to a 2015 Physics Review paper, a published patent application, and a pitch deck.
After 20 years of funding by the US Navy, they are looking for $30 million and 3 years to resolve the remaining scientific questions. They are under US ITAR restrictions, so the work must be owned by US entities.
While the current deep learning hype is strong, the models are still either primitive (e.g. gradient descent) and have little to do with how brain works, or they are too complex (e.g. spiking neural networks) and operate on a wrong level of abstraction to be of any use. Neuroscience made some great progress collecting tons of empirical evidence in the last decades, but there is a large disconnect between research communities. Fund a small team of generalists that analyze heaps of neurophysiological data and reverse engineer the brain.
Traffic patterns. Ahh, just kidding. No really, traffic software. The world is tied up in traffic. Mainly stop lights are are archaic and need such a large improvement with how it detects traffic and stand stills.
Have you ever sat at a light and wondered why it was still red when no cars were coming in any direction?
This software is lousy, it should have object recognition, sensor tech, but the basic idea of traffic lights haven't been updated in decades.
You want to save time, money and a big sale to the rest of the world? Fix traffic lights.
One thought, the only simulation that prevented all other traffic lights or subsequent streets from bogging down in traffic, was to ensure at that particular time, no cars advanced at a certain speed.
Your wait may be a locally suboptimal outcome to ensure everyone else has a globally less miserable existence.
So as another participant in the system, thank you for sacrificing a little so that we can be better overall.
A little food for thought, while driving in another country with less adherence to the laws, the operating custom seemed to be stop if it is red OR if safe, go. Ironically, this prevailing attitude resulted in fewer accidents as people drove more cautiously.
But isn't this largely a social / psychological problem - I assume around 70% of people who commute spend 70% of their time in front of a computer.. so could work from home 3 days per week, if it was socially acceptable to do that.
We've solved things like this before [ daylight saving is a useful group hallucination ] so working from home 3 days per week, or say working a 2hr offset shift could be packaged in a socially digestible form.
1. Health Data Comparison Analysis: What can we actually use existing health data comparison for. How much data is needed for progress?
2. OpenSpaceMining: Research into how to mine space.
3. New OS solutions for new scenarios: Research into new types of operation systems solving new kinds of problems (ex. Fleet management for drones, automated ships, planes, trucks)
4. Technology and it's effect on our economy. What does technology actually do to our economy, how can we make it a part of the economic models instead of just treating it as an externality
Federal Drug Administration: The current pipeline for getting new drugs approved is fairly corrupt ($ > merit) and extremely ineffective (time taken) that its a major inhibitor for innovation in biotech, medicine, and medical devices. Doing research to figure out how to streamline and automate this process while giving a fair playing field to everyone, make it not cost-prohibitive to get new quality products approved while not compromising effectiveness (filtering out harmful products)
How to fix government. Not how to lobby and play political games effectively--I mean really fix it. Transform it with technology. Make it open, fair, consistent, and participatory.
Philosophers of technology. People whose mission is purely to facilitate a conversation about where we really are and what should be done next, and are given the backing and platform to work on and diffuse radical messages within a context different from existing organizations.
Everyone's chained to a reliance on borrowed ideologies of yesteryear, sometimes repackaged to fit the times. There are small whispers of other ideas, but they take so, so long to enter the public sphere.
Gene therapy - It is known that modified virus vectors can fix DNA. I believe this approach can be used to cure wide variety of diseases related to gene mutations.
A better candy bar. Most of the commonly available ones are far too sweet and filled with cheap filler like peanuts or sugar spun air (I guess they call that nougat or something).
It's like they keep reducing the quality somehow... I can't put my finger on it. I swear candy bars were better quality when I was a kid. I know it seems small... but a really good candy bar would go a long way to improving my world! Disrupt Mars!
I think there should be some sort of subsidized insurance program for hiring ex-convicts. No one wants to hire former criminals, which is clearly a problem since unemployment among them increases the likelihood of recidivism. But if companies where protected against the risks associated with employing ex-cons, then they would be more likely to hire them, especially since ex-cons are very willing to take lower wages.
Deer population control. Reduce deer in the northeast and reduce the primary vector for northeaster deforestation and Lyme disease. Come up with an effective way to reduce the deer overpopulation that doesn't involve shooting or poisoning them
In The west, research ways to control the wild boar/pig population, or find a way to produce a sustainable food source from wild boars that doesn't require owning lots of land.
AGI. It's nearly impossible to get a big company to invest in AGI due to the historically bad ROI. That's also the reason there probably won't be a successful startup in this space for a while. The private sector is not particularly great at exploration where ROI is uncertain. OpenCog is a great start, but it would be nice to have another project in parallel that utilizes different ideas.
Issues closer to YC's core competence, such as innovation and productivity. There's lots of advice out there about how to be innovative and productive, but I think it's fair to say that we don't really understand these areas at a fundamental level. Because the YC partners have so much experience and interest in these issues, they could contribute much more than just money.
"Full" auto-pilot. Soon, battery tech will allow us to fly light aircraft on all electric. If people could go to a mini-airport only for short takeoff and landing aircraft (STOL), and get in a 4-seater that will fly to their destination (no human pilot required), that would be pretty close to the vision of flying cars we've all been promised.
Wherever you look, there's opportunity for sampling and sequencing - in the soil, inside livestock, on hospital floors - that could lead to novel life-saving or industrial applications or at worst, a better understanding of the natural world.
To do this at scale takes capital and resources - difficult for many labs, but not a problem for YC Research.
Enabling technologies for improving quality of life and economic opportunity in the developing world, in a way that doesn't require large capital outflow or aid-based investing. EG, low cost renewable power generation, labor reducing tools for agriculture, water purification and management, sewage handling, computing, local manufacturing, etc.
In line with this, $1-10/day basic income in the developing world would be a much cheaper and potentially more impactful way of testing the concept compared to the ~$30/day minimum for meaningful impact in the US.
Society has accumulated enormous "societal debt" (think technical debt). Most things are inherently broken and inefficient. They haven't changed in centuries.
- Communication
- Education
- Emotions
- Jobs
- Housing
- Clothing
- Health
- Nutrition
- Politics
- Economy
We can't fix these things incrementally. The social codebase is worthless. We need to start over, from scratch.
Buy an issland or a big piece of land. Carefully select people based on a yet-to-determine algorithm (people with as little exposition to society, such as kids), and send them there. Closely observe what they do, and ensure they don't repeat and/or bring with them habits and ideas that we know to be wrong. This will take a while, but at some point we will be able measure how they perform as opposed to the rest of the world. These metrics could include happiness (although difficult to measure), ratio of time they spend doing things they agree with (willing to do, not coerced), energy consumed, etc.
Most jobs are utterly useless. Most people have no idea what the fuck is going on. People don't have a purpose or focus. We really have no clue. Logical fallacies make up MOST of our understanding of reality. We live in constant fear and anxiety, and dedicate a lot of resources to cope with it. We quickly become convinced that money is an end rather than a means. We miss the big picture, we lack empathy, we avoid introspection. We believe in lies, myths, bro-science. We don't have a clue about what habits or foods are healthy for us. Regulations and the monstrosity of the legal system make the simplest thing unachievable by mere mortals. We have technology that's thousands of time more efficient than what we had in the past, yet our lives have barely changed (i.e., the hours of work).
Like it or not, most of your active time is spent at work. We work 20-100 hours a week just to get by. Surely, getting rid of 80% of jobs (at least in countries like the US) while maintaining people's quality of life would a real breakthrough. Yet, I would argue that this is already possible, without the need of any additional technological breakthrough (i.e., AI). Look around, most jobs only fulfill habits and historically relevant needs that we can easily get rid today.
Define a location-based metric for climate change vulnerability analogous to the way walkscore.com has defined and developed a metric for walkability.
A good walkscore makes property more valuable. A valid and widely distributed climatescore could be the most effective way to create political will by making property value accurately reflect future risk.
Related to the basic income project -- questions about the societal impact of technology that the technology or business sector might be hesitant to investigate themselves.
For example, how do social networks, targeted advertising, etc. impact social dynamics and sense of self? Are these impacts positive or negative, and could technology improve them?
On a broader scale, how can we measure and understand happiness/quality of life? It is difficult to improve on factors that cannot be measured -- if the goal of technology is to improve quality of life (which I believe it is), then we need a way of measuring that holistically.
Charter cities. https://www.ted.com/talks/paul_romer?language=en Allowing more people to move to a better place seems very beneficial and might have positive unit economics (sustainable after initial investment).
Elder care. Demographics is destiny and with declining birth rates — especially in the developed world — the population will inevitably be graying. How can we provide them with effective and humane long-term care without crashing our economy or occupying a huge fraction of our labor force?
How can we give people education which maps to real life skills in 0.5x the time it currently takes? I spent a semester of college learning to write a memo...(in 2010).
- software assisted learning plans based on interest and need
- cover personal finance, negotiation, and learning techniques.
Government - How do we change politics, so it is more engaging, transparent, and driven by the strategic (rather then emotional or tactical) needs of the country.
How can technology reach people disenfranchised. How can we slice through the FUD spun by the masters of bad/corrupt information.
Origin of life research: Understanding how life arose on Earth or can arise on other planets would be one of the biggest breakthroughs in all of science.
[Sociology/Psychology] How to effectively promote reason, rationality, and critical thinking among the masses.
machine learning on eeg data. finding functions which attempt to map signals to consistent thoughts. software to pre-process eeg data. applications in mental treatment, adhd, meditation therapy, paralysis victims, communication, security, etc.
Biotechnology - protein engineering, synthesized foods, synthetic biology.
Brain-machine interfaces to treat psychiatric disorders (Depression and PTSD among others), paralysis and movement disorders, and sensory write-in to help the blind and deaf populations.
Research research. Lots of money is wasted on regression analysis driven nonresults and other common errors. Best practices for research especially in the era of pervasive software augmentation are very much in question.
(Inspired by the article recently posted here which pointed out that although a drug is being used off label to treat phobias, there is no incentive for a company to get that use certified.)
Optimization of tax code via machine learning. It should be possible to build an intelligent tax machine that uses enough features in order properly classify and tax corporations like GE or high frequency trading funds in the most obviously and proper way. Simple tax codes are vulnerable to lawyers who complicate them.
Delivering internet (and maybe even have core infrastructure like DNS) via satellites (like Google and SpaceX, but not-for-profit) and allowing citizen funding (like planetary foundation)
something that could benefit humanity on the long term is a wider "operating temperature" range. With the planet's weather getting more extreme, and the long term possibility of finding other habitable planets with an atmosphere that might be several degrees off, a human that feels comfortable in a wider temperature range will need spend far less external energy to accommodate. Consider how much electricity can be saved if people felt as confortable at 25deg C as they do at 22deg C.
Every year, there are millions of people who suffer from life threatening diseases like cancer. The statistics refer to only that single person, who is suffering; but the truth is the close family members and friends also suffer, along with the patient.
Usually, they are left to fend alone with their suffering, as it is very difficult for them to share their true feelings with others, because people who are not in their shoes are not actually able to comprehend them.
How does it impacts the patient's care,well-being and recovery? What can we do more to make society more empathetic towards these people? How does this impact economically overall?
How can HN readers help YC research? The list here is very long, and YC might not be able to deal with the whole set. If some of us can help, do mention this as well.
gamification tactics for useful behaviors. Some good tactics should be found out of the cancerous casual gaming wallet emptying strategies. Maybe time gating can get people to write books, that's what the pomodoro method is right? What other tactics could work? Jingling sounds in the "casino acoustics" genre? They've found a way to addict humans to useless behaviors, lets find ways to addict humans to useful ones!
Combinatorial/High-throughput screening for novel materials. The space is need of innovation, talent, and funding in areas that aren't pharmaceuticals.
A solar-powered oxygen producer. Instead of huge solar farms in the desert generating solar electricity, then transmitting it hundreds or thousands of miles to where its useful with all the losses involved, instead directly generate oxygen from sunlight right there. Doesn't matter where you do that for the planet; once up and running it'll help sustain the whole world like a rainforest does. And it could be massively more efficient than growing plants. And it could work where there's the most sun and the least opportunity for plants to grow - large deserts.
1/ Mobile education for young [and poor] children: there are scarce resources for teaching children <=~6 year old. It is challenging to teach some more complex topics. I am not obviously thinking only about coding.
Complementary to the above: making affordable to have a basic robot so kids can learn automating them in the real world instead of the physical screen. I envision a project like One Laptop per Child but for robots.
2/ Affordable Health devices so you can check you up at home or work, probably with you mobile device.
If there were a drug for motivation, that would be a big win. Many people suffer from procrastination, and it seems to me plausibly implicated in poverty.
AI applied to understanding the hidden relationships linking epigentics and extending human lifespan via algorithmically engineered diet, nutrition and longevity genomics:
The precedent:
Laura Deming: Making a business out of fighting the ills of aging
Statistical modeling of biomedical corpora: mining the Caenorhabditis Genetic Center Bibliography for genes related to life span -
Blei DM1, Franks K, Jordan MI, Mian IS. - http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1533868
Combining the worlds variants of pattern recogition, machine learning and AI with data associated to epigentic and standard genomic experiments and information, we should be able to discover hidden relationships that will enable advancement in extending human lifespan.
Revenue is something we'll need to focus on and that may come in the form of consumer applications that algorithmically engineer their epigenome with data from companies like 23andMe, Navigenics, Ubiome etc. connected to Nutrigenomics including plant compounds, phytochemicals mapped all the way to the consumer through foods connected to apps connected shopping and recipe experiences.
Finally connect to CRISPR and Gene Therapy
Jennifer Doudna, inventor of CRISPR: We can now edit our DNA. But let's do it wisely
Education.
Delete non useful (history, astronomy) Insert useful (logic, ethics, mnemonics(not great but better than history)) If the world has changed over 100 years, but the learning about it hasn't, you're doing it wrong.
Congressional Secret Ballot.
Perverse incentives. Congress used to be able to vote privately, now when a corporation buys your vote as a congressmen, they can verify you deliver what they paid for. "Cardboard box reform"
Better candidate pool
Geniuses don't go into politics, the pay sucks and you can't get much done. We can at least fix the pay. If bad voting, but good candidate pool then still win. If great voting but bad pool, only lose. What use is voting when you have choice of puppet A or puppet B both presented by the same megacorps?
Better voting by better voters.
Voting licenses. If you are stupid enough that I don't want you editing the settings of my OS on my computer, why would I want you editing my democracy? Is not governance advanced enough and complicated enough to require a license to effect changes through voting, just as cutting hair is licensed? Yes, its dangerous, yes if the tests suck you get unfair amplified power to the test makers. However, if you want greatness, you must select for it. You will never having voting excellence if you don't try.
Better voting by better intel gathering.
Why capture a single data point of "who do you want to win?" Choose instead "put these people in order you'd take them." Now you've captured 8x the data from the voter? Now you can combined the ordered lists to get an electorate that everyone kind of likes and few people hate, as opposed to the 51 percent voting to gouge out the eyes of the 49 for idle pleasure. https://en.wikipedia.org/wiki/Condorcet_method
Isolate government from currency.
Peer to peer value transfer. A proper BTC/zerocoin might work here. Artificially low interest rates amply mis and mal-investment causing boom and bust cycles that harm the human psycological ratcheting effect. We prefer slow and always better to fast up and fast down pain.
Cooperation.
More perverse incentives. Why must we all keep our greatest ideas secret, so that no one else runs with and executes to our detriment? If you could make a lower barrier to entry profitable collaboration, or set a standard by which idea givers were commonly rewarded, people may be less secretive. This isn't just a corporate problem, this is a science problem as well. Publish or perish? Well better keep those good research ideas to yourself. How many websites have you been on that desperately needed improvement, yet you kept your ideas a secret?
Science is broken
If half the papers published produce results that can't be replicated, then lots of scientists are fucking up. At least remove the "HARKing" Hypothesizing after the results are known. Publish more of what hasn't worked and isn't interesting to save others from unknowingly going down the same dead ends. Sadly, if the science is so bad as to be unreplicable, then who cares if you publish or not as you were 50/50 wrong anyway… The math could be better, however the idea stands that bad science can and is being done to the detriment of the world. Thinking you know a thing and being wrong about it, is worse than thinking you do not know a thing and being accurate in your estimation.
Undercrowding
Progress comes from humans. Through enough shit at the wall, more of it sticks. We need more humans, preferably the useful kind.
Spreading great ideas
The mormans got a bible in every hotel room, what better book could be equally funded and distributed?
Luckily this can be quite a long list as the world is ripe for improvement.
Where to start? If you knew your history, you would realise that attempts to "reset" the education system generally go hand-in-hand with barbarism or huge social disbenefit. How on earth do you "delete" a whole area of human knowledge without burning books, killing people, or otherwise restricting our freedoms? Because free people will choose to study anything which takes their fancy, and there's no practical (humane) way to stop it, or any reason to.
If you understand evolution, then you know it requires variance and combination. If you standardize education, therefore reducing variance, regardless of your selection strategy, your pool of excellence to choose from is reduced. Therefore standardized education diminishes variance and therefore amplifiable excellence.
We need less jacks of all trades and more specialization. You must choose on their behalf, or give them the choice to choose what to not be good at. When is the last time you actually used historic precedent for a decision? How deep is your actual knowledge of history? And could not being excellent in knowledge and action towards today's problems be superior to overlearning about problems already solved?
No matter what suggestion is made in regards to optimizing the education "system" you will find the naysayers that are emotionally invested in learning "all the things" so i mentioned history and astronomy, i could have closed my eyes, pointed at a syllabus, been as likely to find at random yet another suboptimal topic, and i would have in the end only substituted your pro history comments, for pro whatever else comments.
Better and worse exist. Wasting time exists. If you measure what the average or median or whatever measurement you care to use human, and see what they have used and will use of what they've learned, you will find that what you're being forceably literally by law required to be taught isn't what you need to make a living, or be a good person, or have greatness in your life. Its often babysitting hidden under the guise of progress. How can one support a static educational structure in a progressing and more rapidly progressing real world?
Also, you didn't really think I meant to delete history? I hope. That would be the opposite of education would it not? So as to avoid confusion, lets ammend it to "replace rarely used trivial knowledge for commonly useful fits lots of places knowledge, and specialization." When have you last needed to make a left turn at mars and missed your turn? Perhaps you don't need to know the order of the planets from the sun after all?
There are two medical conditions in dire need for an intelligent solution rather than the barbaric approaches in use today. In both of these cases we have friends and family who have had direct contact with current broken non-solutions. I'll split this into two posts so each can get it's own comment thread. Here's the other post:
The current non-solution is to section eye muscles to re-align the eyes. Yes, they pop your eyeballs out of their sockets and cut (shorten) eye muscles to pull the eyes past the point of alignment in the other direction. In other words, they over-correct a little in hopes that the eyes will end-up aligned (or close enough).
This is, for lack of a better term, barbaric stone-age bullshit. And, yes, this is being practiced at places like UCLA on a daily basis. Desperate parents take their young children there and subject them to this kind of butchery.
Why is this the case? Because young children are not treatable. If you've ever dealt with a 2 to 5 year old you know it is just about impossible to get them to run through eye exercises (or any strict routine) for one minute, much less an hour.
So, "researchers" latch onto desperate parents and offer a surgical procedure most sign-up for, with the requisite exchange of a shitload of money for the work.
Children are silent victims in this scam. The surgery is a failure more often than not. The few cases that might provide results are probably corner cases. In other words, the "doctors" get lucky. In yet other cases some cosmetic results are had, yet depth perception is never created or restored. In other words, the eyes might look more aligned yet the kid never sees in 3D.
I met several adults with Strabismus. On person had seven surgeries when she was a kid. Her parents kept taking her to the butcher because, of course, they would see temporary alignment after every sectioning and kept thinking it would eventually "take".
This is disgusting to me. Children are being victimized on a daily basis and desperate parents are being swindled.
As an engineer I see this as a control system problem. The solution, surely, isn't butchery, it has to be in understanding the control system issues and seeing if there's away to re-train or correct it.
I actually tried to have this conversation with the lead researcher at UCLA about ten years ago when someone close to us was considering having their kid butchered. To say that the researcher was a smug condescending asshole doesn't quite describe the god syndrome I experienced in that room. Of course, doctors don't know the first thing about control systems. What the hell did I expect? Right?
I actually took at shot at trying to understand the problem. I read a book titled "Models of Oculomotor Control" [1] and anything else I could get my hands on. I then developed a crude biofeedback device and we tried it on her, then four year old, kid. This is when I understood just how difficult the job of having a young kid go through therapy could be. It was very hard to get him to focus on anything. And this is precisely why butchery wins out when it gives desperate parents hope.
I thought the technology had merit. I thought I saw brief glimpses of the brain saying "hey, wait a minute, what if I do this...". Yet, this effort could not be continued as his Mom was desperate to fix it and opted to go for surgery.
The kid had over-alignment due to muscle sectioning for a couple of months and the eyes diverged to their prior state within a few more months. Now, over ten years later, you wouldn't know this kid was ever tortured through such a surgical procedure. His strabismus is no different than it was prior to surgery.
It would be good to see real research on this front that focuses on oculomotor control, biofeedback, control system training and ways to encourage the development of depth perception in young kids afflicted by this condition. It isn't an easy problem, because the patient falls under the "herding cats" category.
STEAL IDEAS FROM OTHER LIFEFORMS BEFORE THEY DIE OFF
Which organisms are we most likely to benefit from stealing ideas from? Can we prevent them from going extinct before we get to learn those ideas? Is there a way to capture the ideas pre-extinction?
Even the tools that we now have to change the blueprints of life, we stole from life. Bateria and Viruses are the source of our most powerful DNA editing tools currently.
We are meat. Animals are meat. The animals have all kinds of amazing tricks and tools their meat has developed. We could use some of their tools. If they become extinct, we can't learn well from them any longer.
If you want to take advantage of the millions of years of building with the best tools we're aware of, real evolution over long time periods, and you don't want to wait for a couple million or hundred million years for the next go round of that thing maybe evolving again, then you should try to prevent extinction of the organisms most useful to learn from for our own benefit.
MINIMUM VIABLE EDUCATION
What is the minimum effective dose of teachable knowledge/skill. Input, process, output. ala grammar, logic, rhetoric, trivium fame.
Success leaves clues. The most effective often have little mental tricks they apply when thinking, some are mnemonic, some are turning ideas into locations, some are pretending new ideas are just like old ideas, and applying the old learnings, etc. Find those powerful commonalities.
Stop making kids bad versions of google. We don't need more bad data storage in human minds, we need more of what the machines can't do yet, creativity, love.
If you must make kids bad versions of google and shove facts into them for lossy storage, then at least teach them mnemonic technique and logic first, google-fu, so as to amplify all their future efforts. Do not teach the amplifiers last, let them pay dividends over time. I was lucky enough to have logic tought to me first before any other maths at grade 6, because of an experimental advanced education system. After logic we learned scheme programming. I think those frameworks greatly helped me in this life, as nesting of ideas and good data processing is super powerful in all areas of life.
Better and worse exist, find better, choose better. Do not be a coward and pretend that worse doesn't exist.
summary: teach amplifiers first. Then teach minimum effective commonly useful data processing, then specialize and become great.
Specialization is why we have genders. Dimorphism lets us not over-allocate excellence in a single place at the cost of another place. Great things come from specialists. If the value of skill is in opposite proportion to supply of that skill, then everyone learning the same things, by definition means no one will pay for that skill or knowledge. There's some exclusions to that idea, such as language itself, however the theme of slapping knowledge/skill on a supply/demand curve is a good one. Much of this idea is hammer in Peter Thiels 0-1. less 1 to 1 horizontal iteration and more 0 to 1, no one else is working on this thing kind greatness.
What is minimum effective curriculum before specializing?
What is the minimum viable education?
What is the correct execution order of those skills?
For instance, how important is it to teach kids to write essays, when they actually don't have much useful to say? Maybe you should work on having useful things to say much harder before you learn how to beautifully say them. Imagine that you did the opposite, imagine that you taught effective rhetoric before logic, and now you've got better salesmen of bad ideas. That's worse, not better. Think of the places in the world where women aren't allowed to drive cars. How did that terrible situation arise? Great marketing of terrible ideas.
Rethink the current epistemological approach to medicine and the biological sciences.
-
What's broken:
These fields are awash in a sea of data and knowledge. Much of it unfortunately rests upon a shaky empirical foundation. Reconciling new information with existing information is extremely difficult. Ensuring the integrity of the empirical tree, atop which virtually all new research rests—even more difficult.
The process by which scientific knowledge is shared and reviewed—research papers—is at best an antiquated and inefficient mode of collaboration.
Modern drug discovery is extremely expensive. While advances in bioinformatics and computational power certainly help, they are not a magic bullet. Absent complete confidence that a particular drug will achieve perfect efficacy while not causing unintended consequences, extensive study is required. Studies are extremely expensive to conduct and represent a significant portion of drug discovery costs.
The entire premise of pharmaceuticals may be fundamentally flawed. Expending insane amounts of money in search of magic molecular combinations that target very specific conditions, without complete efficacy, via mechanisms that are often not fully understood, and are burdened with a myriad of interactions and adverse events.
The terminology used in medicine and biomedical sciences—cells, proteins, antibodies, antigens, receptors, viruses, fungi, bacteria, paralysis, inflammation, encephalopathy, neurons, dendrites, axons—all of these aren't actual things but rather human-created constructs purposed to assist us in grasping what we're trying to understand. These constructs, while often quite accurate and helpful—do not necessarily map neatly to what's actually happening, nor do they necessarily lend themselves to understanding complex interactions between other constructs.
Empiricism works great until the tree becomes very deep and complex. Then it's problematic. Especially so if the methods for maintaining and growing that tree are far from perfect. At worst, the entire system becomes a hindrance.
-
How to start fixing it:
This isn't a rant against modern medicine. On the contrary, achieving the best results right now with an incomplete understanding, a probabilistic approach (to steal Peter Thiel's terminology) is the best course of action. Studies for example are probabilistic.
Medicine as it exists today has done an admirable job of making the world a far less miserable and deadly place. However, I fear the current approach may not be suitable for achieving mastery of the human body. The depth of empiricism and complexity of knowledge at that level may extend far beyond human mental capability.
To illustrate this point, assume therapeutic nano bots suddenly popped into existence today, straight out of science fiction. No built-in behavior, but fully programmable at a low level. Using them to cure cancer would be an admirable goal, except for the fact it'd probably be incredibly difficult if not impossible to do considering our current knowledge.
Mastering the human body isn't much different. It might require eschewing the probabilistic approach and replacing it with a mechanical approach. Surgeons largely use a mechanical approach in their work, and that's a major part of why they're so successful. At the scale of bones, muscle, and organs the human body really isn't much different from say, a turbofan engine in terms of complexity. Dealing with systemic disease processes on a microscopic scale however, things become far more difficult. While there are many remarkable successes in this area, there's still a long way to go before achieving complete mastery.
As counter-intuitive as it sounds, it may also be necessary to eschew—or at least cease to completely rely upon—the vast tree of existing empirical constructs we've created. Just as a convolutional neural network isn't aware that the photograph it's uncannily redrawing in the style of Picasso contains human faces, a biomedical application utilizing a similar principle may not need to know what cells are—even as it kills cancer.
Conversely, pretend a bunch of really smart people are locked in a room with just a microscope and an encyclopedic amount of samples. These people somehow have zero knowledge of biology, none at all. After a sufficiently long period of time they'd probably come up with an epistemic model somewhat similar to existing biology, but one that would be completely alien in its terminology, perhaps radically differing in certain key areas—possibly for the better. If they had a way to collaborate from the beginning that was far less cumbersome than research papers, it's probably a good bet their resulting epistemic model would far exceed that of existing biology.
The human body is an incredibly complex product of evolution, and as such it is not easily understood by human minds. We're basically compiler output that's trying to reverse engineer itself.
I'm not suggesting AGI as a solution—just that it would be prudent to apply state-of-the-art weak AI in a fashion that's as decoupled as possible from the current epistemic model, because such constructs impose far too many assumptions that might be very wrong in some fundamental way. Ironically the use of weak AI to achieve mastery of the human body would constitute a probabilistic approach, albeit in an extreme form. If a mechanical understanding did follow, it might be very simple or elegant in nature.
This is a benign tumor that grows around the auditory nerve between the brain and ear organs. It squeezes the nerve to the point where hearing on that side is eventually lost. This leads to SSD (single-side-deafness) which can be a debilitating condition.
One does not realize just how useful having two working ears happens to be. SSD causes two significant problems.
The first is the loss of the ability to locate the source of sounds. Your kid calls for help? You have no clue where they are. Someone calls you at the office? Same thing. You literally can't locate sounds.
The other problem is the inability of the brain to run its "noise reduction" routines on sounds coming in. I know people with SSD who, in a busy restaurant, literally can't hear the person sitting in front of them. The brain is able to process sound from both ears and allow the listener to cognitively focus on a desired source. It's a mini phased array with a powerful processor. Take away one of the two available microphones and the loss of capabilities is significant.
Another potential side effect of Acoustic Neuromas is that they can affect the facial nerve for one half of your face. If this happens, the muscles on half the face go limp and the entire half of your face droops.
Current "solutions" don't guarantee the preservation of hearing, hearing recovery or the preservation/protection of the facial nerve.
The first "solution" is to drill a ~ 1 inch hole on the skull right behind the ear and remove the tumor. Of course, the nerve is cut out with the tumor. So, SSD forever. The hole is filled with fat. A metal plate is used to cover the hole. A stud is screwed into your skull and a little box handed to you.
This box is a bone conduction transducer. It has a microphone and transducer that will vibrate your skull. The idea is that your skull will conduct this sound from one side to the other so you can hear sounds from both sides. Of course, you lost the phased array, so hearing this sound, while perhaps useful, might just add to the noise. And, of course, it probably won't be full fidelity. You get to walk around with a pager bolted to the back of your skull [1]. There are other, less invasive, solutions [2] but none that fix the underlying problems. The BAHA devices mean you have this stud --which looks like the end of a spark plug-- screwed into your skull. Skin infections and other problems are some of the issues patients have to deal with.
Believe it or not there's a whole building in Los Angeles dedicated to the above-described butchery. I believe it's run by a USC professor (or ex, don't remember). I believe the procedures start somewhere around $50K.
The next "solution" is endoscopic surgery. Not as brutal. They drill a 1 cm hole on your skull, go in and chip away at the tumor. Practitioners can be more careful and do their best to avoid damaging the acoustic and facial nerves. Yet, nobody will guarantee this. Complete loss of hearing is very likely and facial drooping could happen just as well. While less intrusive than a huge hole and a metal plate, this surgery does not seem to offer improvements in results. You are still likely to be offered a BAHA bone conduction device as an add-on. My guess is this procedure runs somewhere in the $80K+ range.
The third approach is to use Gamma Ray Surgery. This is commonly used for Cancer patients. The tumor is mapped via MRI. The patient's head is then bolted to a table under the Gamma Ray machine and the tumor's blood supply is zapped. The goal is to kill it and keep it from growing. Due to dose limits the treatment takes several weeks. I don't really know how much it costs.
This approach does not invade the cranial cavity, which is good --you can still go scuba diving. It can be precise enough that the nerves are not touched at all. In other words, some hearing could come back and facial feeling/control as well. Yet, nobody will guarantee any of it. Part of this is due to medical liability laws in the US, a whole other subject. BAHA devices might still be offered.
What is needed for people afflicted with this condition could take many forms:
- Early detection: This is a slow growing tumor. Can it be predicted through DNA or other testing? Is there an easy way to find markers and attack the tumor when it is at the 2 mm stage rather than waiting for the patient to notice hearing loss only to find a 3 cm tumor in their skull?
- Better treatment: Can Gamma Ray (or some other non-invasive approach) be improved/perfected in such a way that the tumor can be killed off more effectively while avoiding damage to the nerves and perhaps even restore natural hearing?
- Better devices: Surely there has to be a better and more intelligent approach to restoring some sound perception than bolting a stud into the patient's skull or shoving an uncomfortable lump of plastic into their ear.
Life is quality x quantity. Quality is far into diminishing returns. Quantity can be 10x'd. The best in their fields die, right when they're most useful. What would Edison, Tesla, Franklin, Bell, Feynman, etc be producing currently if still alive?
Since quantity is literally the hardest problem we are aware of, how shall it be attacked? By charity? Let's be serious, look around yourself right now, point to the things charity created. Now cease the futile exercise.
Solution: 1. Marketing as gates does, however all lives are not worth the same, you, your kin, your friends, they are worth more. Fund the research that saves them, do not dream the silly dream that all lives are worth the same.
2. Profit not charity, i.e. take a look at osfund.co (human longevity, inc) sustain and grow, not charity and run out of runway.
3. LONGEVITY FUND. or "save your assssssssets fund" You can buy the S&P500 why not the biotech index minus the companies solving problems you won't have. Reduced malaria drug allocation, increased cancer and heart disease. Guess what. If you survive malaria, you are still going to die of cancer and heart disease, so lets work on those 40 and 40 percent killers.
I've got 220k words on paper.
1. 1st book, personal excellence, things you can do to Scivive. Get rich, be loved, live forever… *if selling ideas is good enough for Bill Gates, e.g. Gatesnotes.com, givingpledge.org, its good enough for me, and you. Spread the word. This genre seems easier to understand.
2nd book, things that are important, however are rarely actionable in your personal sphere of influence (politics, currency, voting, economics, geopolitics, environment, interesting yet un-actionable for the common man things.
3. Longevity fund. biotech index minus things unlikely to be useful in saving you and your loved ones lives.
4. Companies not a subset of Nasdaq biotech index, too small, or in a different jurisdiction.
5. Biotech/synbio startups.
6. Popularity improvement ala "The Martian" whereby science and tech becomes heroic as kicking aliens in the face has been historically. The world need not hero worship the leather ball throwing and kicking, steel lifting and lowering, non scaling great strategies of 2,000 years ago. Let's make heroic the things that actual work these days. How many humans can you heal? Can you save thyself and thy family?
If anyone wants to help edit these 200k+ words I have on the page regarding life and winning at it, hit me at Scivive on gmx.com
P.S. Spread great actionable ideas (sell SP500 buy biotech index.), fund excellence, build excellence, fund riskier excellence in that order.
Fund me and others interested to continue the ideas at http://ifex-project.org around actionable arbitrary economic modeling through an actor-oriented, transaction-level business protocol that is asset type, settlement-system, transaction type and communications paradigm neutral, defined openly and published via the IETF to preserve common access.
Applications: (1) JIT service and product supply chains (2) redundant service and product supply chains (3) dynamic fiscal/product/service logistics (4) overthrow ridiculously endemic middle-men in B2C/B2B (Alibaba, Amazon, Ebay, Taobao, etc.) who provide poor reputation, settlement, escrow and search solutions (5) make sense of 1000s of available shipping and logistics providers (6) change-ready open source fintech supporting nontrivial (eg. multi-currency, multi-stage) risk/transaction/execution models (7) non-conventional asset/settlement-ready decision-making and BI solution (eg. social credits, environment schemes, etc.)
The plan: get generalist people with background in related disciplines together, churn out something usable on a 24x7x365 multi-jurisdictional basis by real businesses, make some functional examples, let it loose, claw some market share from the incumbent closed source players, possibly derive revenue from the subsequent growth in connected services (FX, warehousing and logistics, legals, multi-jurisdiction, e-government) or just fund them as separate entities.
Note that this meets the "stuff not being worked on by hackers, or that is un-fundable by others" suggested requirement, basically because its slightly too big-idea for individual hackers, and too cut-off-the-hand-that-funds-it for most businesses with stakes in the effected domains. It is also a little edgy for most governments, who would prefer people to default to a centralized fiat asset system that is more taxable but usually very high-friction and change resistant. It is also very connected to the basic income notion, as it should easily facilitate concrete modeling of the benefits of such against arbitrary academic models, including actor-emulation, risk-model tuning for all parties and failure simulations: Chaos Monkey for business, economic, legal and social systems. Quite possible impact in law.
It should also work well for modeling a lot of the other ideas presented in this thread and others (eg. Shypmate et al), so many in fact that quoting them in brief apparently exceeded a length limit.
The fact that so many of these suggestions could potentially benefit from a decently extensible, actor-based approach to economic (maybe 'social') and risk modeling becoming open source standardized (and thence easily neural network explorable) suggests strongly that it is a useful area of research. When the idea was first raised, someone from the IETF energy working group suggested it would be a great way to model electrical power in national grids (and possibly in embedded applications). There are no doubt other real-world economies that would love a reasonable toolkit.
I believe that, if backed by YCR, possible members of an initially rapidly assembled team could include 1x very high level, successful global finance entrepreneur and executive (British; already investing in the area, keen to work together, met personally last week), and 1x pure maths + quantative finance masters (Australian) currently wasting brainpower in a conventional investment house sharing interest in the field. The nature of the problem domain is certainly more long term YCR than startup, so it's hard for any of us to dedicate time at present.
YC Research should fund research on synthetic consciousness based on human models of consciousness. Consciousness research is still un-mainstream enough that there's a common perception that no one has any idea how consciousness works or even what it is, even though we actually know quite a bit:
In particular, the dynamic thalamic core theory of conscious experience (Google it, the author-posted paper requests you don't link directly to it) offers a tremendous amount of explanatory power. From the perspective of engineering a synthetic system, you can build on the model to accommodate any aspect of conscious experience. For example, if you have an "input module" that feeds into a visual recognition network that feeds into a conceptual association network, the conceptual association network can activate neurons in the input module that correspond to the neurons that would be activated when perceiving the visual details corresponding to certain concepts. So the input module becomes a place where multiple simulated percepts of concepts can be activated simultaneously, and then "for free" by the nature of the architecture, you get visual recognition processing of the simulated perceptions combined with "raw" perceptions (or combined with other simulated perceptions). As a bonus, you get language comprehension: after perceiving the visual details of a written word, the system would recognize the associated concept (presuming it had previously learned it) and trigger a simulated perception of the concept. Run this process recursively over a sentence, and the system has a simulated experience which would correspond to the semantic meaning conveyed by the sentence.
The human brain has an architecture that appears quite similar: a specifically layered, recursive connectivity between the thalamus (the "input center" of the brain) and the cerebral cortex (which handles low level recognition and conceptual associations, among other things). Regardless of whether or not the human brain works exactly in the vague way I described: 1. You could engineer a system to perform the tasks outlined above, and the architecture would lend itself to whatever facets of human-like cognition you might be interested in (try it, it's easy). You wouldn't be able to gloss over the details like it's an HN post, but there aren't fundamental roadblocks. 2. Neuroscience research regarding the function of the thalamus and the function of the thalamocortical system is very favorable towards this model (particularly if you factor in the basal ganglia), but more generally--the human brain is nowhere near a "black box". We can't read the brain like a hard drive, but it's plainly not a ball of evenly distributed computational goo where any computation could happen anywhere, anytime. For each individual human, specific patterns of activation in the cerebral cortex correspond to specific perceptions and patterns of activation in the thalamus, which implies... 3. The synthetic consciousness I proposed could be connected to a brain-computer interface (attached to your head) (your head, not mine!) (no I want it give it back) so it can learn to recognize your thoughts (this learning process could be enhanced in various ways; for example eye-tracking combined with a forward-facing camera could help the synthetic consciousness know what you're paying attention to, which adds context for learning to recognize what your thoughts may be about), and because the synthetic consciousness has simulated perceptions in the same region as raw perceptions, you can view its thoughts in real-time as well. It's a white-box architecture.
This kills two massively valuable birds with one stone: you get strong AI (more accurately, you get strong synthetic intelligence; "artificial" starts sounding rude after a point), and you solve the "control problem" because you can use your enhanced knowledge of consciousness to develop systems that have human consciousness merged with synthetic consciousness. "What about the human who's enhanced by the merger with the synthetic intelligence though!?!" The integration with the synthetic intelligence makes the human's brain a white box as well. If you wanted, you could integrate multiple such systems together (to whatever degree is comfortable for the systems). So ultimately the control problem becomes a question of how much we trust ourselves and each other after we can read each other's minds. Yes, it would probably be terrifying initially, but I think we'd get used to it quickly, and there are numerous meditative traditions which might be helpful to anyone who struggled to control thoughts that others found particularly repellent.
After that it's a relatively straightforward recursive self-improvement deal. Eventually you improve the system enough that the biological component is redundant (yay). From there you make sure that whatever you're using for physical presence is generally robust. And if you're not incredibly rude, you might be kind enough to use your enhanced intelligence(s) to figure out how to gracefully share the technology with whatever other humans may want it (because presumably the recursive self-improvement process may happen in a relatively short period of time, and it's unlikely that literally everyone will engage in it at once). You'll have to figure out societal structures that make sense for whatever you all consider yourselves to be at that point. Then dive into advanced physics to figure out how to do whatever you want to do in neat ways, and if you decide to make von Neumann probes, don't send them off until the physics research has tapered off, because you'll probably just be beaten by the newer, faster probes if you're impatient and send a bunch off right at the start.
The timeframe mainly depends on how quickly you can iterate on the self-improvement process. Presumably some of that process is going to involve iterating on hardware (and probably iterating on the process of creating better hardware), but the hardware already exists to make a start. Early speed improvements would come from 1. not having to physically type or use a mouse/touchscreen to interact with a computer (programming at the speed of thought?) and 2. some degree of "telepathy" to help coordinate research. Later speed improvements could come from things like "forking" the synthetic portion to run in parallel for some period of time to learn or do things (can merge "copies" back in afterwards rather than ending them abruptly so that nobody has a bad time). You probably wouldn't want to run the synthetic portion at a rate significantly above your general biological rate of cognition, so that would put some vague upper bound on raw speed improvements until the biological portion is completely redundant.
At any rate, the initial steps are quite straightforward and would make addressing every other issue in this thread go faster. It would definitely be faster to go this route than to try cure aging, for example. Biological aging won't even be relevant at the end of this process, aging by definition involves a superset of the complexity of the human brain alone, and progress on the earlier steps towards curing aging do not increase the speed of progress on later steps towards curing aging.
Since I was old enough to understand neural networs idea, I imagined and I was sure that one day we will build an artificial brain, at first with less sub-components (brain areas) and then with more if not all that we know about with their functions. At that time the only thing I thought stops us from doing it is the computing power.
In the meantime we grew enormously our computing power AND we invented the deep learning and a whole lot of faster AI learning algos yet we did not built an artificial brain. Or there are projects but I did not hear about it?...
It depends on how you model the neurons/networks in the brain. For example, if you do it this way: http://www.nature.com/news/fragment-of-rat-brain-simulated-i... then you'll need a lot more computing power, as they do. But they are aiming to eventually build an artificial brain as you describe.
If you just want to create an artificial thalamus (probably connected to deep learning-like layers, since those are loosely based on the layers of the cerebral cortex), you can start much smaller: https://books.google.com/books?id=VTduCQAAQBAJ&pg=PA1159&lpg...
- Somewhat counter intuitively, it works to fund soft subjects like politics, sociology and governance -- because humans have the capacity to create physical technologies we don't have the social technologies to be able to use without a loss of social harmony. This capacity gap is an important opportunity for vigorous research and invention of new social & governance technologies, which don't just include things like digital forms, and also encompass the notions behind the ways we organize and talk about our societies and their purpose. Without funding to advance the conceptual basis of governance and society, the alternative is to rely on advanced military power to provide a guarantee of stability, which is also necessary, and in the long term alone insufficient to create the kind of societies it works to create.
- Military technology, because hope for the best, prepare for the worst, and protect what's valuable.
- Alternative propulsion, because everything depends on transportation networks. The faster we can turn space transport from "horse and carriage" through "steam age" to "high speed rail", the faster we create more wealth -- to fund everything else.
"Military technology, because hope for the best, prepare for the worst, and protect what's valuable." Research and trade of military technology needs to be guarded and regulated much more than it is now. There is already an absurd amount of tax money redirected into feeding weapon business that is killing people and destroying countries. The current state of weapon business is very wrong and throwing more private funds at weapon research to spur creativity and innovation is likely to grow the problems rather than solve them.
Neural networks with the direct objective of creating a "conscious" animal. Right now AI architectures are focusing on prediction, classification, pattern recognition. And this is very, very different from what nature does with AI. Nature controls animal/human bodies, and a few plant bodies using AI.
If we could push an AI controller to the point of having the intelligence of a cat or a dog, we would be able to make transportation equipment that would blow the wheels of of anything else we currently know. Having equipment carts that walk with you, baby carriers that do the same, and have zero issues with stairs, doors, ... (and yes, the military applications I'm sure are wonderful too). The dream would be that you'd have a mobile robot that can deliver packages. Packages the size of refrigerators and that could deliver them in your kitchen, walking in through your door, walking up the stairs/taking the elevator if necessary, ring your doorbell, ... Not to mention the industrial applications. Such robots would enable reconfigurable factories : the robots doing the work simply walk over to their new positions.
Even that would just be the beginning.
Then suppose we get there, can we go further ? Flying animals have extreme advantages over our airplanes. What if you could design a 4 seater airplane with legs and wings like a duck ? A large airplane that could reliably land and take off on a runway shorter than the length of the plane in strong crosswinds. And yet have that plane capable of maintaining flight for hours, far exceeding the endurance and efficiency of helicopters.
What I'm saying is. A lot of limitations of our current state of the art vehicles are effectively a control problem. Building a legged robot is not a problem at all, yet it would have massive advantages. Building a winged robot is harder, but certainly within the realm of possibility. Yet we don't have them. Why not ? We can't control robots like that (see youtube).
Wealthy interests lobby governments to do what is good for them, but often at some negative cost for society. The reason they lobby is the bang-for-buck. What if lobbying is used for good. What if instead of billionaire gives to charity, billionaire lobbies government so that charity is not needed anymore?
Better map-reduce jobs on immutable pre-sorted, partitioned data and related genome sequencing techniques. It must be Plan9 and Go, not Hadoop or anything that begins with J.
No, really. Like Golang - compiling to native code, have a thin layer of proper abstractions, a-la Inferno, reusing OS ABI to dlload and call everything, no Windows support. No VMs - x8664 is a pretty good VM itself. Also learn to stop worrying and love a good OS (the way Golang does).
Mostly functional, with immutable data, which makes really difficult things (runtime) simpler.
The brain. It's probably the most complex machine in the universe. If someday we could make artificial grey matter, or neurons, it would change everything. For AI, for health, for everything.
I’d have said AI safety, but it’s already on your list. Excellent.
I’d also have said basic income. Good to know you’re on it.
Meta-research. Academic publishing is obsolete, and knowledge should not be paywalled. How many researchers already make use of https://www.reddit.com/r/scholar or http://sci-hub.io/?
Bacteriophages. Antibiotics are running out, while phages have been tested by the Soviets, and are available as an over-the-counter treatment in certain countries.
Open-source infrastructure. How many ‘unicorns’ were only possible because of years of prior volunteer work? See @nayafia’s recent post: https://medium.com/@nayafia/how-i-stumbled-upon-the-internet...
Feynman’s and Drexler’s molecular nanotechnology. Smalley muddied the waters, and the National Research Council dropped the ball. We can do better than just nanoelectronics.
See also “Ask PG: What Is The Most Frighteningly Ambitious Idea You Have Been Pitched On?”, paying attention to the Yudkowsky Ambition scale: https://news.ycombinator.com/item?id=4509934