Philosophically, the "thundering herd problem" actually CAUSES congestion.
There are 2 ways to handle it.
1 Stop the thundering herd.(make all the clients do something different). That may make things worse. Congestion in networks is usually exponential. You can't fulfill a request so you repeat the request, it can't be fulfilled so it repeats. You can add a random delay at the client end but that is just The US governments answer to the debt problem, it kicks the can down the road in the short term but it will almost certainly come back and bite you. Mathematically it is very easy for this scenario to become catastrophically exponential when a threshold is reach
2 Stop the congestion (make the handling process faster or add processes)
The system already has a cache to handle this but if its not in the cache it doesn't help.
There needs to be an extra request cache exclusively for congestion scenarios. The existing cache request is already doing some processing, so extend that to route "thundering herd requests processing". This second cache does a bit more processing as well.As each new request is routed to it, it checks itself to see if this requestor is in the cache and removes it or overwrites it. It should never contain more than one entry per client.
When no more editions are made to this congestion cache (or the rate has slowed significantly) then the requests can be forwarded and processed via the original cache system.
Under this configuration, the congestion does not become exponential and should only delay the thundering herd requests. All other requests will be handled as per normal.
Once the original cache has the information there is no need for any thundering herd requests to be routed to the congestion cache.
Some clients will encounter some delays but not all and only on the "thundering herd process".
Well one other aspect of this research springs to mind.
Most "bug zapper" design are wrong. It should consist of a light source and a single grid perpendicular to the light source rather than surrounding the light source.
The light source should also be constantly varying to ensure the insects' tilt (and hence their circling behavior) will also change the radius of their circles.
>Insects use the moon as a celestial compass cue to navigate, and mistakenly use artificial light sources instead
I think that was one of the theories being investigated by this research.
The paper demonstrates that it is the actual light (of the moon or stars or sun or artificial source) they use to orientate themselves in the horizontal plain BUT that is not navigation. With a "tilt" in their orientation they will fly in circles around the light but this tilt also causes inefficiencies in their actual flight mechanism so will cause erratic directional stability as their flight path rapidly changes their spacial relationship to the light source.
Whatever their navigation imperative (heat, cold pheromones,smell, sight, sound) will be affected by this spacial relationship instability.
Realistically this war is being fought politically in the worlds media. The fight is for public opinion.
So the orders of the IJC are about war.
The hair splitters may want to change the nature of war to "genocide" by one side rather than condemn BOTH sides in the conflict.
"but we can only take evidence from XYZ and only on matters ABC"
It seems to me to be a case of the UN wanting to have a seat at the table. They have become more & more irrelevant in matters of real importance or conflict. This ruling is pretty meaningless regardless of what your beliefs are. Its just more media circus.
We can argue about statistics and false testimony who is right and which side is credible but we are just adding to the noise.
Most terrorist organizations & actions are the direct result of one group having no political power. It ceases when that group obtains political power.
It seems to me the original attack by HAMAS was simply to ensure that there is more war and that war can be fought in the worlds media. This ruling and the whole procedure is just more of the same.
Wars usually continue to be prosecuted till one (or both) side loses the political will to continue. Does anyone here really believe Israel will lose that political will, regardless of world opinion? Hasn't in the past.
Just like Russia.
Buy a home "cryovac" machine like food saver.
Prepare cooked meals such as pasta sauce, stews, curry etc in meal size portions. Cryovac and put in the freezer.
To use them put, still in their sealed bags in hot water and bring to the boil.
You can do things like boiled, mashed rosti potatoes. Frozen beans, peas, broccoli etc can be portioned out into separate bags (either at use time or in advance).
It doesn't matter what combination you choose to use, they are all in separate bags in the same pot of water.
The range of meals is only open to your imagination.
Trust Crisis for AI?
What after we see the board/CEO of one company fired/changed apparently for allegations of lies or manipulation that nobody is clear on?
If Dropbox derives data from user data by scanning said data, then that "derived" data is no longer "users data" it is Dropbox data and can be shared.
It may only be statistical in nature and not related directly to individual users but isn't that exactly what training data is?
Isn't that how it works?
That can be shared to train AI models can't it?
Its not lies its hair splitting.
No its called unethical behavior and has become the norm for big tech.
In fairness to Sam Altman and OpenAI, all credible reporting I've heard (and signifcantly Kara Swisher's work) has shown no issues of AI safety or lies in CEO communications to the Board, but rather broader concerns over the direction the CEO and board felt appropriate for OpenAI.
I've no dog in this hunt. I'm not partial to either Altman or OpenAI. And I have considerable reservations over where this Brave New World may be taking us. Or whether there is any credible option to stop riding this merry-go-round, no matter how unattractive the destination(s).
The DropBox behaviour described is only one in a long, long, long line of trust-violating practices by tech firms.
Too often the claims of unethical behavior are knee-jerk, baseless, and speculative. In the example the author cites, Dropbox only sends data to OpenAI when the user explicitly tells the app to engage an AI-related feature, for example to summarize a document. Yet the backlash seems to assume that they're scanning and uploading people's documents en masse, even though there is no evidence of this.
Unethical behavior definitely exists in AI companies. Personally, I'm agnostic about whether it's higher or lower than the base rate of unethical behavior in the general population. Anyway, if we're going to talk about bad behavior, we should use specific examples with cited evidence, not fear-mongering.
> Personally, I'm agnostic about whether it's higher or lower than the base rate of unethical behavior in the general population.
I don't think that AI companies are less trustworthy than our industry in general, but I absolutely think that our industry is far less trustworthy than the general population.
AI companies engaging in the wholesale harvesting of web data to train their models, though, was a particularly egregious violation of trust.
Who is driving the company (companies)? Certainly not the board, when it can be ousted on the whim of one person.
Might as well just get rid of the board (all boards in this organization) as any decision made can be overthrown by a bit of publicity and ego.
Perhaps it about getting rich at the expense of any ethics?
Those who work for this company should bear this in mind, they are all expendable!
The reason the CEO could oust the board like that was that after the board fired the CEO, 720 of the 770 employees of the company got together and signed a pledge that they will all leave unless the CEO is reinstated and the board fired. This was the purest example of labor democracy and self-organization in action in the United States I have ever seen, my takeaway is rather the opposite of "they are all expendable". They showed very clearly that what decides the future direction of the company is the combined will of the employees.
1. A group of highly skilled people voluntarily associating with each other
2. in an organization working on potentially world-changing moonshot technology,
3. born of and accelerated by the liquidity of the free market
4. with said workers having stake in the success of that organization
is very American. We should ponder the reasons why, time and time again, it has been the US "system" that has produced the overwhelming number of successes in such ventures across all industries, despite the attempts of many other nations to replicate these results in their own borders.
> This was the purest example of labor democracy and self-organization in action in the United States I have ever seen
This was not an anonymised vote giving the chance to safely support what you believe free from interference... this was to declare for the King or the Revolution knowing that if don't choose who prevails, you will be executed. It becomes a measure of which direction the people believe the wind is blowing rather than a moral judgement. Power becomes about how a minority can coerce/cajole the impression of inevitability not the merit of their arguments.
I will be curious to hear the inside story once the window of retribution has passed. Unions hold proper private ballots not this type of suss politicking.
It means the board was out of alignment with their employees and their decision was unusually unpopular.
If the board had communicated better (no 2 or different reasons) or had hard evidence (the claim was deceptive behavior) - it would have been different.
I expect everyone who reads HN accepts that execution is 70% of the battle.
It’s very unsurprising for employees to have made the choices they did, given the handling of the situation.
I don't think it means that the board is useless, but rather that their power has a limit, as all leadership positions do, and the way the fired the CEO has crossed the line too far.
Useless in the sense that they showed themselves to be incompetent with their failed coup and had to pay the price. As the saying goes, “if you take a shot at the king you had better not miss”.
Call me a cynic but this must be the reason behind those 700 employees threatening to walk away, especially if as it seems the board wanted to “slow down”.
I don’t think it’s so much of a “we love Altman” as “we love any CEO that will increase the value of our equity”.
People go to work to make money. If they’re lucky, they also like they’re work. It’s ok to want more out of work. Work certainly wants more out of you.
Non-profit just means there are no stakeholders, meaning that revenue must be used for the organization's expenses, which includes the compensation of employees and growth. Many non-profits also benefit from tax-exemptions.
Assuming anything more than that about the nature of non-profits is just romantic BS that has nothing to do with reality.
Also, there's nothing wrong with wanting to make a profit.
Why are you only impressed if the power of employees comes with high cost to their career? When employers exercise their power capriciously, they often do so without any material consequences. I'm more impressed with employees doing the same, i.e., exercising power with impunity.
I'm pretty sure they weren't thinking of how to impress fatbird on HN when they were deciding which boss they like best, the old one or an unknown new one.
They were about to laterally transfer to a new organization headed by the CEO they revered, with promises of no lost compensation. The only change would be the work address. Hardly a sacrifice sanctifying a principled position.
It's a non-profit. It should have no value aside from the mission. The only reason the employees mobilized in their whining and support of a person, whose only skills seems to be raising money and cult of personality, seems to be so that Altman can complete the Trojan horse of the non-profit by getting rid of it now that it has completed the side quest of attracting investors and customers.
Which owns a for profit. They would have been left with no employees, no money, no credits, and crown-jewels already licensed to another company. On top of that spending life in minority shareholders lawsuits.
And you believe they could’ve achieved their mission with almost zero employees and presumably the board (well… who else?) having to do all the work themselves?
> non-profit by getting rid of it
So the solution of the board was to get rid of both the non-profit and for profit parts of the company?
There was nothing to bluff. The employees would've been hired by Microsoft and OpenAI the company would've had nothing but sourcecode and some infrastructure. They would've gone down quickly.
What would that have accomplished? The old board members would still be in charge of the same thing they are today: absolutely nothing. They overplayed their position. That's all there is to it.
That's one interpretation, and I'm sure it is the one they had. In my opinion they failed to enforce the charter. Had they been effective at enforcing the charter they would have not folded all of their power to someone they view does not enforce the charter. There is nothing they could have done to fail more.
>Had they been effective at enforcing the charter they would have not folded all of their power to someone they view does not enforce the charter. There is nothing they could have done to fail more.
How specifically do you suggest they should have "not folded all of their power"? Are you saying they should have stuck to their guns when it came to firing Sam?
In any case, it's not obvious to me that the board actually lost the showdown. Remember, they ignored multiple "deadlines" from OpenAI employees threatening to quit, and the new board doesn't have Sam or Greg on it -- it seems to be genuinely independent. Lots of people were speculating that Sam would gain complete control of the board as an outcome of this. I don't think that happened.
Some are saying that the new board has zero leverage to fire the CEO at this point, since Sam showed us what he can do. However, the new board says they are doing an independent investigation into this incident. So they have the power to issue a press release, at least. And that press release could make a big difference -- to the SEC and other federal regulators, for example.
I guess this might have seemed like their best option if a public statement would be a breach of NDA's. That said, I still wish they'd just have some backbone and issue a public statement, NDA's be damned, because the question of upholding a fiduciary duty to be beneficial to all of humanity is super important, and the penalties for a violation of an NDA would be a slap on the wrist compared to the boost in reputation they would have gotten for being seen to publicly stick to their principles.
>I guess this might have seemed like their best option if a public statement would be a breach of NDA's. That said, I still wish they'd just have some backbone and issue a public statement, NDA's be damned, because the question of upholding a fiduciary duty to be beneficial to all of humanity is super important, and the penalties for a violation of an NDA would be a slap on the wrist compared to the boost in reputation they would have gotten for being seen to publicly stick to their principles.
I see what you're saying, but unilateral disclosure of confidential information is not necessarily in humanity's best interest either. Especially if it sets a bad precedent, and creates a culture of mutual distrust among people who are trying to make AI go right.
Ultimately I think you have to accept that at least in principle, there are situations where doing the right thing just isn't going to make you popular, and you should do it anyways. I can't say for sure if this is one of those situations.
Well they might have as well announced that they are dissolving the company and the outcome would’ve been similar. If that’s not “overplaying your hand” I don’t know what is
There is literally no evidence to it. If it is about the charter they could directly say it was about the charter rather than using strong language when firing and going silent.
So no one, not even their new CEO they picked knows the reason of firing and this reason has been cited as truth by HN multiple times.
I wouldn't say overplayed as much as badly played because they underestimated
how much their CEO's had fortified his position. I find the situation pretty dire: we need more checks and watchmen on billionaire tech entrepreneurs, not less.
I wonder if they ACTUALLY would have landed at Microsoft. And then, would they actually have their TC post tender met there? And would they continue working on AI for some time, independent of reorgs, etc? All of them? A place like MSFT has a TON of process and red tape, the board could have called their bluff and kept at least 40%.
Please explain why you think they should have let them walk and what benefit to either OpenAI or any of it's partners / customers there would have been in doing so?
It feels like you are saying this out of spite rather than with a good reason - but I'm open to listen if you have one.
They could, or could they? Formally boards don't run companies. Would this scenario - besides ridiculous - be legal? Say they let 99% of the company walk. Who is qualified and responsible to do whatever needs to happened after that happens?
I've always like the metaphor that companies, especially large ones, are like an autonomous AI that has revenue generation as it's value function and operates in that interest and not in the interest of the individuals that make up the company.
Ah if only it were that simple, truth is a company, especially old ones, have a value function that is in flux, and this may include revenue generation but more often than not this is completed or replaced by replaced by some variation of political clout acquisition: it makes them unkillable even with a negative balance sheet.
Just the whim of 1 person? Just the ego of 1 person? 700+ egos it seems. Ultimately, what is an organization if not the people who make it up?
Regardless of what actually happened in this Mexican novela, whoever was steering the narrative - in awareness or not - led everyone to this moment and the old board lost/gave up.
> Who is driving the company (companies)?
There's probably not a single answer that covers all orgs at all times. I don't know if the core of what happened was due to a fight between "safetyist and amoral capitalists" or "egomaniac and representative board" but it the end it was clear who stayed and who left. Ideas are only as powerful as the number of people of gets persuaded. We are not yet(?) at a moment where persuading AGI is sufficient, so, good ideas should persuade humans. Enough of them.
In the end, social constructs drive such decisions, not computer algorithms. Having independent boards adds more friction and variety to major decisions. That does not mean that there is no way to influence them or change boards. It’s how the world has always worked. It’s disconcerting to those who think about such things in simpler black and white terms, but there is a higher logic to the madness.
Why do the hackers support Sam? He's the least hacker-y of the major players at OpenAI. Seems like more of a fundraiser, marketing type with little technical skills. Hackers traditionally support technically competent people.
Reading through comments I felt that at least superficially many in HN thought this was a fight between "business exec(s) and technical/qualified scientists" as if any organization can exist based purely on technical (as in technological or academic) expertise.
If anything, this novela should tech the importance of politics in any human organization.
That demand on HN was an unfounded gut based reaction that you typically expect from a mob. People were saying stuff like the board was mentally deficient for doing what they did. How likely is it for the board to be actually dumb to the point of actual stupidity?
If you think about it, the board being actually mentally challenged is an extremely low probability but that was the mob reaction. When given so little information they just go with the flow even if the flow is unlikely.
Now with more articles, the mentality has shifted. Most people don't even remember their initial reaction. Yeah It looks like everybody on this thread has flipped, but who admits to flipping? Ask anyone and they'll claim they were skeptical all along.
People say The hacker news crowd is different, better... I'd say the hacker news crowd is the same as any other group. The difference is they think they're better. That includes me... as I'm part of HN so don't get all offended.
As bad as the mob mentality is on HN, I think it is much worse on Twitter. I got a bunch of upvotes here on HN for linking the charter and pointing out that terminating OpenAI as an organization would be quite consistent with the charter. I didn't see the same point getting much play on Twitter at all.
I think the fact that upvotes are hidden on HN actually helps reduce the mob mentality a significant amount. Humans have a strong instinct to join whichever coalition appears to be most powerful. Hiding vote counts makes it harder to identify the currently-popular coalition, which nudges people towards actually thinking for themselves.
I think maybe what you're trying to say here is that HN tends to be anti-capitalism.
Personally, I'm pro-capitalism. As a capitalist bean counter, I feel that it's important for OpenAI to uphold its "primary fiduciary duty" to me and other humans, as described in its charter: https://openai.com/charter
Battery Technology is a bit of a handy distraction. The only solution is to move the "Duck Curve" to later in the day. So use batteries? well yes if you use them at the household WITHOUT solar.
If you have solar + batteries you just increase the duck curve
You can use centralized batteries but that probably means major upgrades to the distribution infrastructure.
If the power companies would charge according to cost of supply (so ALL power is cheaper during the Duck Curve period) so you can charge the home battery during "duck Curve" and use it ONLY during the evening peak.
But most power companies have multiple generator technology and use power generated by "cheaper" generators to subsidize the costly overnight or non renewable generation.
Sort of...
But Firefox is open source, so anyone can contribute. Mozilla accepted money from google to "just make google the default search engine".
But they also accept contributions in programming (its open source after all) from google, Microsoft & others employees.
The google/Microsoft contributors whether working after hours or as part of their employment have some influence in the direction of Firefox.
That may be deliberate or benign (a really good direction by google, in a workers mind, will bias their thinking in other projects and visa versa)
That is one disadvantage of open source. A foundation can't avoid that unless you setup like Greenpeace and don't give the riff raff any real say.
Firefox users are "captured" by any adds on google search even though they may advertise chrome.
People that comment here are interested, most users of browsers ARE NOT. It is so much easier to use the default. Google knows this, Twitter knows this.
One of the more brilliant ideas in Firefox was containers, chrome could not implement them because of architectural constraints but rather than build on that beautiful idea, Firefox went in a slightly different direction that Chrome & Edge could follow and still protect their advertising revenue.
Just because anyone "can" contribute doesn't mean every patch is accepted or that product decisions are delegated to corporations with competing products. Firefox still bears responsibility for its own product decisions, not to mention how firefox is marketed.
absolutely correct, but Firefox takes notice of ideas.
They can be influenced by what is proposed as fixes or contributions.
"Everyone knows" that Firefox may break some websites, just ask google.
Has anyone ever come across that pretty regularly?
Firefox has become a follower rather than a leader and I think the influence of contributors has done that.
It is no good for Firefox going in a direction that a good deal of their contributors wont go. The users be damned!
Power of attorney- no real help, you are dead.
Depending on your country....
When you die your estate "owns" everything and only your estate can operate it.
But any property in joint names AUTOMATICALLY becomes the property of the joint owner and IS NOT subject to any restrictions
What????
In most jurisdictions, the law says the executor of your will can operate on the will, its contents etc, but there are strict legal requirements and it is NOT automatic. A court usually has to OK it.
In Australia it is a probate court. It takes several weeks to make sure the will in question is the real and most recent.
As a general case, the courts allow the estate to pay for things like funerals etc.
If there is no will then someone has to apply to the courts for administration.
Your "next of kin" is a Hollywood dramatization and in Australia has absolutely no actual power to do anything.
Usually the "next of kin" will actually do the organizing to inform the executor (if the executor doesn't know)
So the important things to put in your "In case of Death file"
Location of will
Name of Executor
Burial insurance details (if you have any)
Location of a secondary file for the executor with email account and website account passwords, Bank Account details, Insurance policy details
Location of a secondary file for the executor with email account and website account passwords, in short everything the executor needs.
I also have a document containing the "order of my life"
This is how I have organized my life.
Whose name the cars/boat/golf buggy is in, what insurance companies, Utility companies for power water and services, clubs I am members of, things like roadside assistance etc.
How I buy and sell my equities and shares, who I am paying hush money to and who is blackmailing me from porn hub.
Basically all the minor things I do in everyday life.
If you are uncomfortable with this topic, if nothing else investigate an "Enduring Power of Attorney" (it may be called something else in your jurisdiction).
This expires with your death but it allows the holder to make medical and other decisions should you become unable to make decisions for yourself (dementia, coma, stroke accidents that leave you in a vegetative stake)
(In Australia a spouse has that right automatically but if your spouse is unable to make that decision (died in the same accident, has dementia etc) then you NEED an Enduring Power of Attorney).
Your children do not have the right and your executor in the future can take legal action against them)
This is way more important than a "in case of death file"
I know the post above was long, but I must mention the contents come from personal experience.
One thing I found, sadly, was that people I love and trust including family often become greedy and grasping when there is inheritance at stake. Bear that in mind when you make a list with access details to things like bank accounts.
You may not believe it but that is what will happen to your family.
I like to think it is a reaction to grief, rather than an innate selfishness.
There are 2 ways to handle it.
1 Stop the thundering herd.(make all the clients do something different). That may make things worse. Congestion in networks is usually exponential. You can't fulfill a request so you repeat the request, it can't be fulfilled so it repeats. You can add a random delay at the client end but that is just The US governments answer to the debt problem, it kicks the can down the road in the short term but it will almost certainly come back and bite you. Mathematically it is very easy for this scenario to become catastrophically exponential when a threshold is reach
2 Stop the congestion (make the handling process faster or add processes)
The system already has a cache to handle this but if its not in the cache it doesn't help. There needs to be an extra request cache exclusively for congestion scenarios. The existing cache request is already doing some processing, so extend that to route "thundering herd requests processing". This second cache does a bit more processing as well.As each new request is routed to it, it checks itself to see if this requestor is in the cache and removes it or overwrites it. It should never contain more than one entry per client.
When no more editions are made to this congestion cache (or the rate has slowed significantly) then the requests can be forwarded and processed via the original cache system.
Under this configuration, the congestion does not become exponential and should only delay the thundering herd requests. All other requests will be handled as per normal.
Once the original cache has the information there is no need for any thundering herd requests to be routed to the congestion cache.
Some clients will encounter some delays but not all and only on the "thundering herd process".