I have taught a course on quantum computing a few times, mostly to CS students who have no background in quantum mechanics. The way I proceed is to
* First introduce classical reversible computation. I model it using linear algebra, meaning classical n-bit states are 2^n length binary vectors, and the gates are 2^n x 2^n binary matrices acting on theses states. Exponential, yes, but a faithful model. The critical feature here is that you already need the tensor product structure. Rather than some unique feature of quantum.
* Introduce probabilistic classical computation. Now the states/vectors have real entries in [0,1] and obey the L1 norm (the critical feature). Similarly, the gate matrices.
* Now, argue that quantum computing just requires the same linear algebriac structure but we (1) work over the complex number field, (2) norm is L2.
The reason I like this development is that it takes at least some of the mystery out of quantum mechanics. It is not a strange model of computation, completely divorced from classical. Just a variant of it, that happens to be the one the universe runs on.
Peter Shor does discuss classical computation in two lectures, but from just the notes it seems detached from the rest of the course.
I find this notion that the UK is a police state quite hilarious.
Ours is not the country where the police bring guns to seemingly every minor dispute and fairly often draw them. Ours is not the country where police kill you for resisting arrest, stand by while your kids are murdered in a school, or seize your money out on border roads without needing cause. Ours is not the country with a toxic plea bargaining system that throws the book routinely and a 90% conviction rate, or prosecutors who run for election on promises to be ever tougher. Ours is not the country with three strikes laws, death penalties, tent prisons run by fascist antiheroes, rampantly profiteering private prisons, corrupt local sheriffs, newspapers getting raided when they investigate the local police chief, Stand Your Ground and SWATting.
Yeah. We overpolice people being rude. It matters when you have the equivalent of one fifth of the population of the USA crammed into a country a bit smaller than Michigan.
Batshit crazy is clearly subjective, right? Try looking at things from a different perspective.
The problem with the "I've got nothing to hide" argument is it's not "you" who decides what is "right" or "wrong". The entity doing the "spying" determines what is right or wrong. "You" might think "x" is ok, however the "spying" entity may have the opposite view. And it is the "spying" entity's opinion that matters, not yours, because it always them that have the power and authority in determining what is "right" or "wrong". Moreover, definitions change on what is "right" or "wrong".
"Employee benefits? No no, they aren't employees, they are contractors who operate with a dramatically higher degree of autonomy... Except when we fire--er, unilaterally sever relations, because they are being autonomous in ways we wouldn't accept from employees." /s
Drop the quadratic term, equate to 0, and you get an approximate solution for x for the next iteration x_{n+1}:
x_{n+1} = x_n - Df(x_n)^{-1} * f(x_n)
Like this it's the Newton method. The problem is that the Jacobian Df(x_n) is a matrix of size k x k, and inverting it may require O(k^3) work.
So, pretty much all schemes are based on approximations of the Jacobian.
Gradient descent for example replaces Df(x_n)^{-1} with a scalar a, so it's O(1) work to "compute" it:
x_{n+1} = x_n - a * f(x_n)
A method like L-BFGS tries to build a relatively cheap approximation to the inverse of the Jacobian during iteration, resulting in O(k) work to compute:
x_{n+1} = x_n - P * f(x_n)
Other methods may exploit sparsity of the Jacobian, or solve the linear system Df(x_n)z = f(x_n) only approximately for z using e.g. iterative methods.
Note: in optimization problems, the function f is typically a gradient of a cost function, and the jacobian is then the hessian of that cost function.
Apropos of anything else, J&J's handling of this has been despicable. Creating an entity specifically to pick up all of the related liabilities, picking up those liabilities and then (exact count may be wrong, but it's very close) literally filing for bankruptcy within the space of three days.
The Texas two-step, as it's called.
> In 2021, the company spun off its liabilities into a new entity called LTL Management under a strategy called the “Texas two-step.” That legal but controversial approach allows a solvent parent company to protect its assets by creating a subsidiary to hold its liabilities — and then having the new company declare bankruptcy, as LTL did just days after incorporating. Critics complain it’s an abuse of the bankruptcy system and allows companies facing massive litigation to dodge corporate responsibility.
> The Third Circuit Court of Appeals in Philadelphia dismissed the bankruptcy filing in January, finding LTL was not in financial distress and had “no valid bankruptcy purpose.”
> But hours after that case was formally dismissed in April, LTL filed for bankruptcy protection again, this time with a higher settlement proposal of $8.9 billion.
The fact that this is legal is such a mockery of consumer protection law.
What would be the greater loss of life and harm to innocents- allowing China to continue, or world war 3? Because war with China will definitely devolve into a global conflict.
We couldn't even get India or Brazil on board with sanctions against Russia for invading Ukraine, and nothing short of military invasion will stop the CCP from doing what it does. Devout religious faith is entirely antithetical to the CCP rule- both in the idea of a higher power from the state, but also China's history wrt. the boxer rebellion and other movements.
Minimise investment to maximise profits, which you take, whilst running the minimum service required to avoid contract penalties or publicity so adverse you lose your contract.
Then threaten to go bust and get bailed out by the government when the infrastructure starts to break a few years later.
> a much greater danger of moderated models reinforcing politically correct groupthink
Where's the danger in that? For decades now, I've listened to people whine about "political correctness" trying to stir up some fear that it's coming for our favorite jokes or whatever, yet somehow it never arrives. Instead, what happens every time is that the "politically correct" stuff just ends up boring and bland and nobody engages with it. It has no cultural significance. The incessant complaining about "political correctness" is the deathcry of an older generation that no longer controls the hip lingo and must instead follow someones norms.
Political correctness is not dangerous. It's just boring.
And, once again, people seek to add additional, unnecessary steps and abstractions to perfectly fine processes, because of contrived bullshit ("copyright!")
Stop abstracting input away from output. Attribution here is a fools errand built on overbearing, unjust, and misinterpreted concerns over copyright, the relevance of which have yet to be proven in a legal sense. This is going to give The Man yet another way to suppress actors they don't like.
Repeat it with me, folks: YOU CANNOT CONTROL THAT WHICH HAS BEEN RELEASED TO THE PUBLIC. If you do not want things (people/their agents) to perceive your works, then keep them private.
At the risk of sabotaging stuff I've worked on, it's _astoundingly easy_ to glue together Y.js, prosemirror, and a little bit of express.js to get a distributed Notion clone with conflict free resolution.
Like I have something that I got working in 3 days with multiple pages, so links between things, and cursor syncing (most of this work is provided by wonderful people with full-featured libraries!). There's some trickiness with how Y.js actually works (in particular map ID conflict resolution meaning you don't want a dictionary of page names to pages but UUIDs and then some sort of manual revision checking....). But you can get that and probably with a bit more work get ~instant search and the like. All with free offline capabilities.
Of course then I stared at this and was like "I like notion because it looks nice" (I personally don't care about the AI stuff or the database stuff), so just went and paid for notion for my family usage.
When Chelsea Manning was arrested, Daniel Ellsberg would remind everyone that the stuff he leaked had a higher classification than whatever Manning leaked.
The Most Dangerous Man in America: Daniel Ellsberg and the Pentagon Papers is a great documentary about him.
Fun fact: Mike Gravel was an Alaskan senator. When he heard Daniel had the papers, he convinced Daniel to send them to him. Then Gravel went on a filibuster to prevent funding of the Vietnam war, and he chose to read the Pentagon papers aloud in the filibuster. By doing so, he ensured that they would legally be accessible to the public, because they were now part of the congressional record.
Because humanity and capitalism's incentives are just wrong. Bottling companies like Coca Cola and Snapple have long switched to plastic bottles, and externalized the cost to the environment.
My recommendation would be to tax negative externalities and redistribute all of it as a UBI to the people of the country. Simple and effective, but apparently the governments have been moving way too slowly.
What's worse is that the governments perpetuate a lie to the public, making them think they can individually make a difference. In the case of plastic the lie was "recycling", when in fact the plastics were simply shipped to China, who dumped them in landfills and rivers.
But the government tells the individual that they can't have a plastic straw or bag. It's all there to distract the individuals from banding together and demanding the costs be imposed on the corporations which put out metric tons every day. I write more about this phenomenon here: https://magarshak.com/blog/?p=362
And it's not just the bottling companies. It's all the packaging. It's the clothes using synthetic fabrics like polyester, which generate microplastics flushed in every wash. And so on. Convenience is when you'd rather have a one-time-use spoon shipped from China, than wash and re-use a spoon. Your ancestors re-filled containers.
If we made it more costly for these companies, they'd long ago have researched biodegradable alternatives.
I don't know why you base your statement on the assumption that the industrial highly concentrated and unhealthy (for the animals definitely) style is the basis such a discussion should be based on. You lower your bar for your side of arguments substantially by pretending the line is there.
You even do it without the shadow of a hint that the base you use is entirely artificial and unnatural. No, animals should indeed not be raised like that. But I find such arguments dishonest at best.
I have yet to find a discussion where any significant (or really any, but let's assume difficult people exist everywhere) number of meat eating people argue that the current state of how animals are raised in industrial-style facilities using what you mention is how it should be.
If they let cows roam free - like what indeed happens in plenty of places in the world - then they use free energy stored in grass. Which is the much better food for them anyway.
It's almost like a lot of people completely forgot that North America once had tens of millions of big grass feeding animals roam what now is the US (https://medium.com/@davbunnell/once-there-were-50-to-100-mil...). I'm not saying that could happen again at that scale, but it is always (and I'm not exaggerating by using that word) missing from the argument of the "anti meat" side, whose arguments read as if having a lot of grazing animals is something unnatural and can only occur because we raise them artificially and that getting rid of them is necessary to help the climate. If that simplistic argument were true Buffalo Bill did something for the environment by killing the huge herds - which I don't is not what people mean.
A magazine called The Economist should understand why we use almost all of our arable land for farming. It's because that's the cheapest way to do it.
We could grow all of our food in greenhouses and return 99.9% of our farmland to nature. We don't do that because that would be more expensive and cheaper alternatives make growing wheat & corn in greenhouses highly unprofitable.
Conversely, if suddenly 75% of the arable land came onto the market, it would drive the price of land down dramatically, making it profitable to farm it less intensively (aka more cheaply). In other words, we'd still farm about the same amount of land, but we'd use less fertilizer and herbicide and the price of food would go down. Those are awesome results, but land usage would not be significantly different.
> the larger the group, the less likely the group is to be atypical.
Not true (unless I misunderstood what you were trying too say). The probability that a sample will be atypical under the null is exactly the significance level, and does not depend on its size.
What does change with size is that the distribution around the statistic gets more concentrated, meaning your threshold moves to the left, towards lower effect sizes.
Which also means that for smaller sample sizes, it is harder to reach significance, unless you're dealing with a fairly large effect size.
No my argument is the definition of a moral imperative…
Kant sort to solve moral questions using pure reason. Culminating in his famous categorical imperative, also known as the universality principle: “It is our duty to act in such a manner that we would want everyone else to act in a similar manner in similar circumstances towards all other people”
You have some how confused this with pragmatism, and while Kant could be viewed as quite pragmatic, taken to its extreme, a pragmatic position would be: if you are hungry and human meat is available, then eat it.
Just to play devil's advocate for fun... what follows is not a sincerely held position on my part, just enjoying the dialogue—
I don't want a lion to eat me. But I don't have a moral complaint against the lion for wanting to, trying to, or even succeeding. I'd be pissed, of course, but my objection is purely that I don't want to be eaten at all, period. There's not a moral dimension to it at all.
I also don't want YOU to eat me. But unlike the lion, I DO have a moral objection against you, on top of the simple not-wanting-to-be-eaten-at-all thing.
In conjunction with that, I find that I also have a moral objection against MYSELF, when it comes to the thought of ME eating YOU. So, although I don't know for sure yet what the basis of that moral objection is, I have learned that there does exist one. And, I can conjecture that whatever the foundation of that objection is, that's the same basis for my moral objection against you eating me.
But, whatever that basis is, it doesn't seem to apply to a lion who wants to eat me, because I don't morally object against the lion. So, although that doesn't guarantee anything one way out the other just yet, perhaps it's justified that I don't feel any moral discomfort with me eating the lion[0], while absolutely having immense moral discomfort with you doing so, or even really wanting to. The situation, morally, with the lion seems to be a different thing, morally, than you eating me or vice-versa.
([0] This is speaking in the moral abstract, anyway—as it happens, there's not really enough lions left in the world to waste eating them, but that's separate from the question of whether it's moral to do so in general, if situational stuff like species extinction and other ($n)th-order effects weren't a factor).
The article says that the wage penalty applies to people with "strong regional accents," and that Southern is the fourth most likely accent for job seekers to try to suppress, with New Jersey being #1. Framing this as specifically Southern inverts the actual finding, and is just culture war chum.
I run some forums, some of them are quite large. Recently the big increase in scraping by the search engines (Bing has had the greatest increase) caused me to question why.
It used to be that the cost of scraping came with the benefit of being search engine listed which drove traffic, but that feels less true than it used to (for a lot of reasons).
But now the cost of scraping doesn't feel in the favour of a website.
Scraping and bots are for search engines listing, technology tests / experiments, advert / audience measurements, brand protection, IP tracking, copyright enforcement, screenshots for links on other websites (i.e. Facebook), Pinterest linkbacks, training of LLMs (my hypothesis on Bing's massive increase), spam, etc, etc.
With the search engine value lowered by less traffic, yet a solid community still growing via word of mouth... the rest of those things offer no value to me or the community. So I asked the community, what do you want to do here? Leave them all? Ban some? Ban all? Some midway thing?
Almost unanimously the community (who fund the costs by donations, and at least 30% of all traffic and costs were known to be associated to bots) chose to block every bot.
So that's what we've done.
We've blocked every major hosting and Cloud ASN, or put a challenge up to the few known to be proxies (i.e. Google Data Saver), and we've blocked hundreds of bot user agents, we've blocked requests where no Accept header was present where it should be, we've blocked TLS ciphers that aren't modern web browsers — I looked at requests by Python, Go, Curl, Wget, etc... and blocked everything that obviously differed from a valid browser.
In the end we blocked about 40% of our traffic, and so far not a single real human has said (and it's a tight-knit but large community with lots of ways of contacting me) that they've had any issue at all.
We appear to have reduced our traffic and associated costs, with no loss to us at all.
I need to handle (for work) a graph with 40 million nodes and more than 130 million edges.
As expected, networkx couldn't handle more than a million nodes so I had to search for python libs which might handle that much data.
Having a graph with several million nodes isn't just some edge case, social graph for instance grow way faster than anyone could expect.
So I do totally agree that there is a gap between popular graph libs which handle very small graphs and real life libs which need to handle way bigger graphs.
Btw, thanks for the work you've done with 'nodevectors'.
PS: I'm not criticizing either networkx which is very handy and quite good when prototyping a solution.
I think "why do you have the right to know?" is a decent answer.
"What level of trust do you think you've earned?"
"What do you want to do with all this information?"
"How will this make things better?'
"Don't you have more important things to spend time and money on?"
"Is this really the best thing the government should be working on right now?"
"How does this help any of society's ills?"
"Will taxes need to be increased to manage, store, and analyse the mountains of information that will be created?"
"How long will you store messages between me and my daughter about her contemplating suicide, and will those messages prevent her from getting a government job later in life?"
"Does the storing of all this data come with a responsibility to make it accessible to defend people who have been accused of a crime?"
"My wife works in law enforcement, will she have access to this trove of data?"
"Will there be a process to remove data from this trove that may be libelous or cause undue harm to an individual or company?"
"Is the government responsible for the complete chain of access to this data, or are there private companies involved that have to be trusted not to sneak a peek?"
"What specific issue, currently facing today's society, does this solve?"
Our two closest relatives are the chimpanzee and Bonobo.
The Bonobo is far more interesting. They have a dynamic society, where sometimes it's matriarchal or more balanced. The wiki article is a fascinating read:
> Although a male bonobo is dominant to a female in a dyadic interaction,[46] depending on the community, socially-bonded females may be co-dominant with males[47] or dominant over them, even to the extent that females can coerce reluctant males into mating with them.
Personally I think humans are far closer to Bonobos than chimps in our societal structure. We should allow more dynamic structures to arise, and have more sex along the way.
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
It is also possible that a host key has just been changed.
The fingerprint for the RSA key sent by the remote host is
SHA256:uNiVztksCsDhcc0u9e8BujQXVUpKZIDTMczCvj3tD2s.
Please contact your system administrator.
Note that the SHA256 present there matches perfectly the one github send. If you don't remember the very first time you connected to github you also had to accept the key. The warning above shows up because the server is saved as a different RSA, for the SSH client it seems that someone setup a server for github but has a different key, which could mean someone is trying to trick you into connecting into the wrong server. This could also mean that github changed their RSA key which is why they published this article.
GP works for Adobe, and Adobe's bread and butter are the professional creators who would love a world where there is hardware DRM on your eyes and you can't even see their creations or a likeness of them without paying a royalty (and one to "rent" the memories of the visualization, not to "own" the memories like we do now). While I largely agree with you, the GP post is exactly what I would expect from an Adobe person.
`ollama run codellama:7b-instruct`
https://ollama.ai/blog/run-code-llama-locally
More models uploaded as we speak:
https://ollama.ai/library/codellama