"Finnish-American software engineer and hacker Linus Torvalds has as estimated net worth of $150 million and an estimated annual salary of $10 million. He earned his net worth as the principal force behind the development of the Linux kernel."

Source: https://www.therichest.com/celebnetworth/celebrity-business/...

There is limited rational or ethical reason for one human being to accumulate so much centralized wealth. There is being rewarded for hard work / skill / innovation and there is exploiting people to become rich, and the line between those is very hard to distinguish when talking about individuals making more than 9 figures off of the endeavor.

It is also worth considering you don't make billions operating a nonprofit or making a valuable software product (or any product, in general). You make billions founding or investing early into a multinational corporation that amasses more wealth than many nations.

And that is where I think the conversation breaks down - people take personal offense to conversations about wealth and equality but we aren't talking about people wielding the influence of themselves as individuals, we are talking about people who control more assets than some nations, that can potentially have financial influence on the world stage greater than the combined influence of millions of fellow humans.

First: yes, there are ethical reasons to accumulate so many resources: because some problems are incredibly hard to solve and need a lot of resources to even start to solve.

Second, it is not on the people who have earned something to prove why they should be "permitted" to keep it; it's on the people who believe it should be taken away from them to justify that. Saying "it's a shame that there are billionaires" is a complaint that some people are wildly successful and shouldn't be, or that they shouldn't be permitted to keep what they earn. There is justifiable resentment and offense to a conversation that starts with a thinly veiled demand for justification, rather than a presentation of justification. The burden of proof is approached the wrong way, and nobody seems to discuss that.

(I'm not interested in quibbles over the word "earn" here.)

It's not a shame that there exist billionaires. It's a shame that there exist people with nothing. The former doesn't need fixing; the latter absolutely does.

This is a great example of status quo bias.

You believe that the onus of arguing ethics is on the people who want to change the status quo when really, we should start by thinking about what society ought to look like from first principles. And then you absolutely do have to give arguments for why you want any inequality to exist in the first place, let alone billionaires.[0]

> It's not a shame that there exist billionaires. It's a shame that there exist people with nothing. The former doesn't need fixing; the latter absolutely does.

While I'd agree that the latter problem is more urgent, the former can't be dismissed so easily, either. For one, the existence of billionaires (and upper-range millionaires, really) is pretty antithetical to one of the foundational principles of democracy, which is "one person one vote". It gets even worse when you consider the possibility and practice of inheritance.

And besides, some people might argue that if we didn't have the former problem, it might be easier to tackle the latter -- because when there is less inequality, there tends to be less focus on wasteful jockeying for social status overall, and then people might be more inclined to help those that are below themselves on the social ladder.

[0] There are good arguments in favor of some inequality, but the degree of it is pretty important.

Some people are worse than others. Some people are less useful to society than others, though that doesn't make them worse. Some people, given large amounts of money, can do amazing things for the world. On a whim, I can think of Elon Musk. If he weren't allowed to have his fortune from Paypal because inequality is unethical... who would have gotten that money? A government? What would they have done with that money? And why is government/the public always the best decider of how wealth is spent?

Your argument may be reasonable in degrees: Who needs $500 billion dollars? Should there be a legal requirement to invest that money in the public at a certain point, if it isn't taxed directly? I don't know. But I believe inequality is ethical and necessary.

Both, for two main groups of reason.

The first one is that when the available positions in society are highly unequal, then equality of opportunity may simply not be enough for a fair outcome, as who ends up in the comfortable and high-status position ends up largely being down to mere luck.

The second one is that (at least as long as families and inheritance are a thing) the two are linked anyway. Inequality of resources and status is the leading cause of inequality of opportunity, after all. You basically have to either go Brave New World or increase equality of resources in generation N in order to make serious progress towards equality of opportunity in generation N+1.

And we agree that a lot of it is about a matter of degree -- I've said as much in my original post, after all :)

I'm not in favor of inequality, but I can't disregard the fact that it is naturally occuring in most realms. From geography, to agriculture, to education, to skill, etc. If there is one principle that I think is true is that in nature there is a complete lack of equality, yet there is balance.

I understand that equality is an ideal that we aspire toward. But it almost seems arbitrary and forced when you consider that systems don't inheritely have equality built into them.

Sorry if this doesn't make sense, just trying to spark conversation.

Interesting challenge, taking a look at the need for inequality based on first principles. Let me attempt my hand at it.

If you look at a capitalist economy as an optimization mechanism, large wealth concentrations are sources for large jumps in the solution space that escape local maximums. Explaining why:

Start by ignoring the comfort providing aspect of wealth. At these levels, wealth is more about power to shape society than it is about personal comfort. On the personal comfort scale, you get diminishing returns as you spend more.

In terms of shaping society, a billionaire has more power than a millionaire. Whereas, for a large project, a millionaire must convince other millionaires to share the risk, a billionaire may throw himself on a tangent all on his own.

This single-mind behaviour, when compared to group-mind behaviour, produces solutions that are a lot different from the status quo. In terms of optimization, they are larger jumps in the coordinate space.

Now, these large jumps in the solution coordinate space may be beneficial, may be dangerous, or altogether innocuous. Most are innocuous, some are wildly positive (Tesla pushing the auto market to move towards electric traction), some are wildly negative (Koch brothers capturing government).

The question of whether the final balance is positive or negative is an extremely difficult question. We're talking about very rare events in the optimization of very complex systems.

So, I have a very nice, useless theory :-)

Given the current massive wealth inequality, one really simple, effective, and ethical way to help people who are poor and suffering is to redistribute wealth. We don't do that because our political and cultural systems disproportionately represent the interests of the rich.

It's reasonable to say that as long as there are poor and suffering people, as long as we don't all have access to health care and education, as long as there are causes for the common good, there may be an asymptotic upper limit to the amount of wealth a single person can ethically have, when that wealth could be going toward improving other people's lives instead.

The closest analogue from before would have been inventing something clever, making a proof of concept, filing a patent, and chaching you never have to work again and have your own Senator. And of course that never happened.

Linus obviously puts in a hell of a lot more work than that, but the simplified version is a thought experiment.

Anyway, nobody begrudges someone who had a good idea and did the world good, having it made and living a good life. But I think there's two elements that turn things sour- one is the influence of money in politics, and the other is as you mentioned that some people have nothing. If nobody was starving and we got money out of politics, it would be easier to applaud overnight billionaires with no reservations.

As for the "Hard Problems", it's businesses or government that tend to tackle those. The billionaire investor-philanthropist funding a derring-do venture out of his own pockets ala Musk is not the norm.

Allowing people to become billionaires is a hideously inefficient way to solve hard problems. Most billionaires hoard their wealth, or spend it on the hard problem of ostentatious display of wealth. Billionaires who solve hard problems: Bill Gates and Elon Musk. The rest of that wealth goes on things like solid gold soap dishes for superyachts - my neighbour worked at a shipyard, and a customer had them rip out merely gold-plated soap dishes in all the showers.

If even a fraction of that money funded university research departments, say, we'd get far more problems solved.

One word in, and already the same inverted burden of proof problem as the comment I replied to.

> people to become billionaires is a hideously inefficient way to solve hard problems

It's a good thing that's not the reason, nor is any reason required, and that's not your choice to make. And, also critically important: it's a great way to solve specific hard problems that aren't the ones most people are focused on.

Over the course of my lifetime, I expect to donate 99+% of my lifetime earnings towards specific charitable causes that are not something that the vast majority of people care about, though I believe they should.

> If even a fraction of that money funded university research departments, say, we'd get far more problems solved.

University research departments are a passable way to solve some problems, motivated by the ability to attract grant money and publish papers. Many problems are far more efficiently and quickly solved when motivated directly by producing a product or service people will pay for.

Again, the problem is not that we have millionaires and billionaires. The problem is that we have people with nothing. Let's fix that.

Good on you.

No, you'd get more grantsmanship. It's an indicator of how hideously inefficient and misguided our university research programmes are that the word even exists.

If you take into account the historical development of the world you basically argue for the natural state of affairs. The stronger or more powerful person is generally right and free to do as they please and the weaker have to bring forth good arguments or band together to become stronger to assert there „rights“. This is certainly the way it used to be and often still is but we humans have come up with societal arrangements to change this because we realized it is ineffective - there are simply more fights and less cooperation otherwise.

As you alude to in your last statement you agree that humans have basic rights and there needs should be met. However, regarding the effectiveness of social arrangements, do you really believe a society with a handful of super rich and a broad majority of have nots is going to be effective and cooperative? If you think about the natural state of affairs, who is more powerful, the many or the few? While it’s certainly not easy to organize large groups of people, it‘s possible and happening as seen with Trump and Brexit. These are movements which in my opinion have a strong attitude of „sticking it to the man“ and are a direct consequence of inequality. So I think if you somewhat follow your own logic to the end the billionaires actually have the burden of justifying their wealth to the general populace. Otherwise you would be mixing and matching rights and the natural state of affairs in a way that contradicts your own argument in the first place.

I'm not sure if you are being facetious or you generally don't see the link here.

(Leaving aside that we already have the money to help people who have nothing and are just really terrible about actually doing it for all sorts of reasons that would not be solved by having more taxes.)

Of course it's easier to get a gun then to become a billionaire. But billion dollars is closer to nuclear energy than to pistol.

I'm not sure if billionaire-control laws would be beneficial, but I understand why there are at least a few positives.

With YC and other incubators acting as a filter for where investors should put money, it matters more which zipcode you were born into than your overall intelligence and creativity.

People talk about bitcoin, but it's the only 'startup' that regulation didn't prohibit people from investing in. Facebook should have made millenials/gen x rich. Github should have made developers rich. The list goes on and on.

Right now, we're spending a lot of money trying to figure out how to get people out of these ruts. (Free housing seems to be helpful.) Creating more ways for people to get themselves into them seems ill-thought-out.

On the other hand (to pile tangent upon tangent), something like UBI would neatly fix the "problem" that investor accreditation addresses, because it would mean that even the poorest people would have a cushion in case their investments don't work out.

No it doesn't. It's pure nanny statism. The government has no business telling people what they can and can't invest in or spend their money on. Yes startups are risky. So are lots of things. Why should the wealthy have no restrictions on them, but the poor do? This is literally "class warfare".

Private companies are a very specific kind of company—they're supposed to be a temporary condition, a way for companies to not have to go through the effort of going through disclosure, underwriting, etc. right away, because they're still just two guys in a garage and don't have the resources to do so. Private companies, as a special case, don't get to participate in all the things public (i.e. "normal") companies do. Private companies are to public companies as minors are to adults.

In your phrasing, you could say that government also "tells people what they can and can't" eat, in the form of the FDA. But, just like this case, it should be clear that the FDA never goes after someone for eating a poisoned cucumber; it just goes after the company for selling poisoned cucumbers.

(The DEA, on the other hand, is pure nanny-state-ism.)

In your poison cucumber example I think this would qualify as knowingly poisoning someone else which is clearly illegal and we do not need the FDA for that. As long as the cucumber is labeled as "poison" I have no problem with people selling them.

To be clear, the strict purpose of the FDA is to ensure that for things-where-you-can't-easily-tell-what's-in-them, like food or drugs—but which have ingredient labels (or names that serve as common-sense ingredient labels), the thing must actually contain what it's labelled (or common-sense expected) to contain.

The FDA looks for poisoned cucumbers, because the FDA's job is to define "cucumber" and stop people from selling something as a "cucumber" if it's not one.

That's about poisoning; but it's also about high levels of weird things like potassium, that are fine normally but which people don't think of as being in a cucumber and might interact with the drugs some people take. It's also about products that don't contain the right ingredients (i.e. why Cheez Whiz isn't Cheese Whiz.)

And, for drugs, it's about the name mapping to a registered claim of a particular effect, such that manufacturing pills that don't have that effect means you can't call them by the registered name. (Otherwise you could sell "ibuprofen" with 2 nanograms of ibuprofen per cap. The ingredients list wouldn't be a lie, per se; it would still contain ibuprofen.)

"In your phrasing, you could say that government also "tells people what they can and can't" eat, in the form of the FDA."

"The FDA doesn't tell you what drugs you can and cannot take."

The FDA's actions do effectively stop you from having any source to buy a poisoned cucumber, or homeopathic ibuprofen, or whatever else, where said source is a US corporation. But only because it's going after those corporations. It doesn't care what you do.

(In a world where the FDA did care about such things, we'd probably have "drug tourism" laws the way we have "sex tourism" laws, where you could be prosecuted for drugs you used while abroad.)

That's not what the FDA does (the DEA, OTOH...)

The FDA deals with how drugs can be marketed, not what people can take.

If that's all it did, the FDA would be an ideal system. However, that's not true; the FDA also deals with which drugs can be marketed. Effectively, the FDA is indirectly preventing citizens from taking experimental drugs, whether they have a terminal illness or not.

IANAL, but I'm fairly certain it's actually the solicitation part that's illegal. In theory, the company could accept it without an issue if they could prove that they didn't solicit at any point. In practice, there's no way to do that reliably, though.

It's not leftover morality laws; it's that monitoring commercial gambling to protect against exploitation of the poor is a really hard problem, one we haven't really solved. Some states choose to allow it despite this, and then deal with the consequences of increased poverty. Most don't. (And then they're forced to allow it in anyway by native on-res exceptions.)

This is also why such commercial gambling establishments as are legalized, are heavily regulated. For example, see the Nevada state laws on gaming fraud: https://www.leg.state.nv.us/NRS/NRS-465.html#NRS465Sec070

Those laws aren't there to stop people trying to cheat the house. They're there to stop the house from exploiting people by rigging the game.

And we all know no poor person in these states ever gambled their money away! Or, alternatively, they did gamble it away, but since gambling is illegal, they were also cheated mercilessly and have no recourse and no place to complain, since being poor and telling police you participated in illegal activity gets you an express ticket to jail.

Gambling is viewed as entertainment, both by the government and the people participating in it.

As entertaining as startup life can be, there's a very big difference between a risky investment and gambling for entertainment value.

I might advocate for shutting down gambling, credit cards and getting the government out of student and home loans, but if we think those actions should be are allowed, investing in small companies should be also.

Why not reintroduce prohibition while you’re at it? It is not possible to force your opinion on other people like that, it only drives the behavior underground, which is a breeding ground for crime.

Its just the Maggie meant what she said about a shareholding democracy

And even Sweden has a strong retail shareholdings they even have national organisation that looks out for their interests

Anyone can invest in publicly-traded companies. The law prevents unaccredited (i.e. not rich) investors from investing in private companies.

A company has to go through a number of processes, such as disclosure and underwriting, when it goes public. These processes effectively make it a much safer investment vehicle, amenable to purchase by pretty much anyone. A company that hasn’t gone public hasn’t gone through these processes, and therefore might just be a huckster’s sham.

For a look at what it looks like when private companies are allowed to raise money from the public, don’t look at other countries (which have similar laws) or at Bitcoin (which nobody but rich people even understands.) Look at early Kickstarter, before they put in more stringent disclosure requirements; and look at Indiegogo, which still doesn’t have such requirements. Especially look at projects that started on Kickstarter, were kicked off for not meeting the disclosure requirements, and then went on to raise money on Indiegogo. You know what percentage of those are scams? It’s not small.

Early US stock market investment, before the requirements to IPO were made stringent, look a lot like Indiegogo. Anything and everything was up for investment, without examination. Many people did lose their livelihoods.

The first thing to come to mind for me, besides taking a test, is for people to show they can make it as a blackjack card-counter -- seriously, wouldn't an official investor-accrediting poker league, even, do a better job of sorting out or educating the marks and freeing up much more of the population to invest? If we have to deal with this problem by prior restraint at all.

Something that also should be legal (with reasonable discrimination). I would choose a different metaphor because this one doesn't work in your favor.

But I don't believe that we would benefit, on net, if you could buy heroin in the corner convenience store without a doctor's encouragement to do so. The number of people who would be able to safely manage their own treatment, would be outweighed by the number of people who would become newly addicted due to increased access and the aura of illegitimacy being removed.

See http://hf-implode.com/index.html#lists for a partial list.

Even the famous Marc Mezvinsky's notoriously "well supported" Eaglevale Partners fund has shutdown up after losing 48% of investors funds.

There's really no reason to agitate for people to invest their savings directly into a single series A round. Capital pooling and diversification is a key function that VC provides investors.

For instance, until now ICOs have been able to raise money without a prototype in hand, this I believe will change.

What you’re effectively saying is, if you’re stupid enough to invest in a scam then you should be allowed to. The establishment shouldn’t put barriers in your way if you want to give your money to a hoax.

There’s something to be said for that. If you’re not capable of managing your own money then maybe the con artist will do better?

Personally, I’d prefer to give to a charity with a decent track record but it’s your money. You presumably don’t need nannying.

You are fundamentally saying all ICOs are a scam, otherwise your comment doesn't make much sense. Because that's false but it's true there are a number of scam ICOs, it's also true we can lower that number. If you read my comment, you probably noticed the part about the need for the ecosystem to mature also.

And I'd like the ICO industry to die. Quietly. In its sleep if possible, not with all guns blazing.

Why? Because the Satoshi network is one of the most important societal developments of the last 20 years (if not longer). It's important because it made more progress in the field of trust than we've made since, I don't know, the Templars probably.

And what do we do with that gem? ICOs. ICOs, whose trust model consists of anonymous people asking for money with vague promises about a non-existent future. It's a travesty. It's like drawing a smiley face on The Scream. Das is nicht einmal falsch.

And you're right. It's not worse than opioids or gambling. But really, is that the bar? Shouldn't we be building on the progress rather than pissing on it? Wouldn't it be better if we were working on the trust problem for exchanges? On the oligarchy problem of mining pools? Taking the flaws in the current network and fixing them?

So yes. I'd like the SEC to step in. Maybe the old guard can help the rotten branch wither so the tree can flourish. I hope.

If you want another cesspit, can't you dig it somewhere else?

Point being the government deciding how to "protect us" is based on arbitrary criteria at best.

It would be neat if money was spread even more evenly. But the costs of the intervention to try and wangle it would probably outweigh the benefits.

That isn't to say that taxes shouldn't be higher or benefits higher. But even Sweden and France have billionaires. They just have their wealth in foundations and in Luxembourg.

Change "neat" to "critical", and I'll agree with you.

My gut feel is that the US is close to another revolution. All those less-educated white males with several automatic rifles each are not going to like it when they find out their beloved Donald (oh look, another billionaire solving hard problems!) makes their lot worse instead of better.

Its an intentional misdirection through disinformation, but rich and powerful men have been adept at it for centuries now, they learned well from the examples of the French, Bolshevik, and similar revolutions on how to make sure the impoverished masses are focusing their discontent on anyone but the ones truly responsible for their plight.

If you wield power as long as you keep the poor fighting amongst themselves they can never organize against you. See how Occupy Wall Street fell apart through mostly infighting and misinformation as a simple example of this effect that isn't exactly contemporary.

The problem is impossible to solve because those who are doing better than the US still have this problem?

Or you buy land in the 80s.

The problem is that billionaires buy politicians, people are too poor to live in western countries despite working, sometimes two jobs, and people with income from work pay for social security and income taxes while billionaires with income from capital gains don't pay relevant taxes and don't pay for social security or their fair amount for health care.

So focusing on billionaires is a distraction. We need to focus that people have meaningful income - especially with the coming AI/robot revolution - and meaningful lives. Capital gains and companies need to be taxed the way people with income from employed work are taxed and need to contribute the same amount to health care and social security.

But this will not happen, because see "buy politicians". There is only one reason that billionaires and celebrities hang around with boring politicians. It's the best investment.

> The problem is that billionaires buy politicians

Is your solution somehow forcing billionaires to not do what their wealth allows them to? If they can, they will. That's why having billionaires is a problem.

This line of thinking always makes me laugh. You don't complain when a doctors makes money by serving hundreds of patients every month right? Then, when someone builds a product/service that is used by hundreds of millions, or billions of people, the natural conclusion is that they will become millionaires and billionaires.

It's very natural, and very logical.

> we are talking about people who control more assets than some nation

How about folks who centralize powers like politicians in high level functions, who have not even been elected but appointed? How do you feel about that? They also can have massive influence on the world's affairs, even if they are not billionaires. I'd be way more concerned about that.

Being natural and logical doesn't make it moral or ethical. I think you're only kidding yourself if you believe that people deserve that much money and power.

> How about folks who centralize powers like politicians in high level functions, who have not even been elected but appointed? How do you feel about that? They also can have massive influence on the world's affairs, even if they are not billionaires. I'd be way more concerned about that.

Nice use of whataboutism. We can protest, resit, or vote for politicians and bureaucrats. A billionaire will be a billionaire unless the economy collapses or they decide to become a philanthropist.

Turns out people want money, and will offer you violence in return for it if you offer enough to justify it.

And who is it to decide? What organization? Money is simply a proxy for utility in society (at least for salaries) based on demand and offer. Do you want to replace that by divine birth rights? Casts? Or cancel private property altogether? So far that's the "less worse" system we have found that does not make people live in abject poverty. And you know what, it's working. Look at WHO poverty reports: poverty is decreasing progressively around the world.

If there is a demand for the use of certain assets without the hassle of owning them, why shouldn't some billionaire invest into that to meet the need.

They have to leave the country with the fear of extradition at some point? They have to give away their excess, but who gets/takes the money? They go to jail? Who enforces/regulates that people are only earning/saving what they are allowed?

It's sounds virtuous to say "Things would be better with less wealth inequality" and I agree that it would be a better situation, but any means to actually move towards that situation gets ugly fast.

Sorry, my cynicism is showing. But we already have the organisation (IRS) in place to regulate people's income, and they already do exactly this job. Yes, we do need to find an impartial way to redistribute the wealth, but we have working models of an array of options around the world. With functioning government, this is really not that difficult.

The trick is the corruption already happened. The money was already too centralized, and has already done damage to the integrity of public institutions. That however is not a reason to abandon such institutions in favor of the monied interests that brought about the corruption, it is a reason to more seriously pursue means to purge the corruption and restore democracy. When that centralization of wealth causes that corruption, that means you need to address that vector in your ideology and policy goals as a citizen.

Ignoring it is just a long term means to eventually collapse the society as a self sustaining growth of both inequality and corruption consumes all prosperity and growth.

The default option is putting it in the government budget. Seems obvious to me.

If we reach the point where this cap-tax is able to pay for the entire government by itself, and we still have extra, then we can worry about where the rest goes.

> Who enforces/regulates that people are only earning/saving what they are allowed?

The IRS.

That's the "ugly" part. And it gets really ugly.

And has nothing to do with taxing personal wealth over 10 million (or other arbitrary big number).

Some of those companies have produced far more than some nations. What of it?

Tons of corporations are worth a billion dollars, many more than individual humans are worth a billion dollars, and the implications and side effects of each controlling such large centralized capital differ substantially because of the rules of business they each abide by.

But why do you think that such an assumption could be even remotely true?

We tried even distribution and found that everyone being poor sucks, so now we let some people get rich first then actually have something to even out.

Very optimistic but seems to be the official reading of the plan.

I mean $150m is a LOT of money. $10m/yr is a lot too. Just to put that into perspective, you could put that in a very safe ETF and make over $1m/yr. You could easily make $10m/yr just off interest.

That's A LOT of money. And a lot of people would be perfectly content with that, let alone know what to do with it. I mean if you can rent the entire country of Lichtenstein for over 12 years, I'm sure you're pretty comfortable. I don't think Torvalds is in it for the money. Measuring success by somebody's net worth is a really poor metric.

We can even refine more. What is success to you? Total wealth? Leaving an impact on the world? Finding yourself? Being happy? Money definitely helps with all of these, but when you have the income we're talking about, how does it help you then? There are diminishing returns. The only version of success it substantially helps you with is if you measure success by wealth. And most people don't actually internalize that definition. So I personally do not feel like you can say Bezos is more successful than Gates. Or that Putin is more successful than either of them. Is Linus less any successful? You can't judge that without knowing what their goals are/were.

You can absolutely SPEND $1 billion. There's no way you can ever APPRECIATE that money.

It's a bit like calling someone from Quebec "French".

Yes, since Swedish is a Germanic language, I suppose. I know a bit of German too, and have noticed that when I read words in Scandinavian languages, I can sometimes guess what they mean, due to their being similar to corresponding German words.

Had been reading on Wikipedia about different human languages and their backgrounds and histories a while ago, and was interested to see that Finnish and Hungarian had stuff in common. They are both Uralic family languages according to Wikipedia.

https://en.wikipedia.org/wiki/Finnish_language

https://en.wikipedia.org/wiki/Hungarian_language

Take a look at Beowulf and good luck trying to understand the original germanic version of English. You won't.

[1] https://groups.google.com/forum/message/raw?msg=alt.language...

[2] https://en.wikipedia.org/wiki/Uncleftish_Beholding

Sorry if this is way too much topic drift from Supercomputers!

https://en.wikipedia.org/wiki/Swedish-speaking_population_of...

And also that there's no standardized definition of what a "supercomputer" is.

My iPhone is not a "supercomputer" even though it has thousands of times more computing power than a Cray 1, which is considered a "supercomputer."

Seems to be a Linux kernel fork with performance optimizations.

EDIT: I stand corrected https://en.wikipedia.org/wiki/CNK_operating_system

  With an anti-science regime in charge of the government, America will only continue to see its technological lead decline.

[0]: https://m.youtube.com/watch?v=MShbP3OpASA&t=39m

That said, why bother? Linux is customizable enough to strip out almost everything extraneous for nodes and has the largest pool of experienced developers, and it's unlikely that any of the other options would have any performance advantage - and less likely that they could keep any performance advantage once it was identified.

RT is all about privileging latency over throughput, whereas in HPC throughput is everything.

The term you want to search for is "OS jitter", lots of papers on that topic.

I'm not saying this is actually a sensible tradeoff for any extant HPC architecture, but it could be. You certainly want the 'OS' that runs the pipeline on your CPU to be real time!

It actually means "latency with a known upper bound". The point of it is to be able to give engineering guarantees - the actuators on the rocket engine vanes will move within N milliseconds of the input voltage from the gyroscope module changing; the robot arm motor will be de-energised within N milliseconds of the laser perimeter sensor activating.

Bounding the worst case can often mean making trade-offs in the best case, and that doesn't make sense in HPC because the hard upper bound just isn't necessary in this environment.

Apple's Xserve cluster was an interesting experiment but it never really went anywhere.

A pity Apple does not manufacture real servers any more. With the old cheesegrater-mac-pros, which were the last thing available after the Xserve death, you at least have the option to put in 4 disks and do a RAID setup, but no remote power/console management or a serial port...

Competing in the server market was very difficult for Apple, their needs are often radically different from consumer or even "pro"-consumer. They realized the couldn't "win" at servers, so backing out was the best bet for them.

The need for servers is there, or otherwise the "macOS Server" would not be in the App Store. It actually is really nice, it provides everything needed for a business: LDAP, file server, mail, calendar, webserver, TimeMachine centralized, print server, ... - but how am I supposed to recommend this to a customer when Apple does not have any kind of HA ensurance (dual power supply, remote management, auto power on after AC loss, proper SNMP support, Wake On LAN), a way to attach more than 4 disks (or a real RAID, not a software one like CoreStorage does, and they cut out RAID management of diskutil anyway) or actually hardware that I can mount in a rack without wasting immense amounts of (expensive) space?

For the small-business scale, Microsoft actually has the SBS Server licenses, which are not tied to any specific vendor and are exactly the same tools used at big companies so a company can hire any competent consultant/admin for management.

Hell, they could simply ask Dell, HP or whomever to sell a specific server as "macOS Server compatible", charge a bit of money for the OS and that's it - people would rush to buy it, and if only to run a CI/CD environment for mobile app building without having to run it on non-DC-grade hardware!

The "Server" software they have is for small businesses and is pretty good for 10-20 users, but beyond that you'd probably use Google Apps, Microsoft Exchange or something more serious. It's a nice thing to have, but it can't compete at an enterprise level.

There was a time when you'd find server rooms with a mix of AIX, HPUX, IRIX, SunOS, VMS, A/UX, and Netware and that was considered normal. In an academic environment you'd have someone running Plan9 or BSD some other quirky thing as well just because.

Then Windows and Linux came along and started to kill off this diversity one system at a time.

https://www.top500.org/statistics/overtime/

Then select "Operating System" and submit, you can see the OSs that used to be run on systems over the past 10 years.

It would be interesting to know what version of Linux is used most often on current supercomputers. The supercomputer here at Goddard runs SLES (Suse) and I believe other NASA supercomputers run that as well.

Basically anything else that operates on their hardware infrastructure is ok. Though it may be very challenging to get a useful supercomputer on very old operating systems.

I'm guessing one of the major factors to decide for Linux is this:

"The licensing cost of a custom, self-supported Linux distribution is the same, whether you're using 20 nodes or 20-million nodes."

If you want to play around, go to http://openhpc.community/ , fire up a few virtual machines, follow the installation instructions, and start playing around.

  - 300 PFlops: Linux
  - 260 PFlops: other 
  - 150 PFlops: Cray Linux
  - 130 PFlops: CentOS
  - 1 PFlop: bullx supercomputer suite (supports RHEL/SUSE/etc)

[1] https://share-ng.sandia.gov/news/resources/news_releases/san...

-----------------

"The semiannual TOP500 Supercomputer List was released yesterday. It also shows that China now claims 202 systems within the TOP500, while the United States claims 143 systems."

My fellow Americans:

Did you ever want to directly experience the transition to being a Tier 2 country? This is what it's like. It starts with the cancellation of the collider in Texas in 1994 (see Tyson's comments on that one). It continues as steady decline through the next two decades, under both Republicans and Democrats, and now it is reaching its culmination in the Trump administration with the replacement of scientists by political hacks. It's not fun. It's quite sad to watch, that's for sure.

The PRC is striving to become a global competitor in as many fields as it can, as quickly as possible. But sometimes they go after the status symbols of achievement versus actual productive achievements. Both the ghost cities that stood initially empty and the supercomputer clusters that lay idle are results of this pursuit of the appearance of being a global player. [Apologies for not providing references here]

Most countries these days (post cold war) don't build super computers for bragging rights; they build them for defined tasks that drive their specification. Economics, budgetary concerns, and even power consumption often trump absolute performance when specifying a machine for a task.

The rise of clustered machines with commodity parts also distorts things since a faster computer (by some measures) can be made by adding more nodes rather than applying any fundamentally "better" technology.

That's not to say that the PRCs efforts are in vain. Focusing on competitiveness has produced results, and politics notwithstanding, the existence of these capabilities will create opportunities for their use.

But yeah, I was bummed about the SSC (especially given the decision's geopolitical context). I also agree that given fundamental research's past dividends, we should be funding a whole lot more of it.

We seem to be flailing in 4 (or 8) year cycles with most of the progress made in previous cycle being wasted or destroyed within the next cycle.

And there really aren't that many supercomputer installations, so any that RHEL does have are probably more for the prestige than the revenue.

This is idiotic. The new admin could well commission new supercomputers for espionage or other military purposes.

The constant politicization of every single aspect of life is going too far.