HNHacker News
TopNewBestAskShowJobs

pdonis

18,511 karma · joined August 9, 2010

Email me at peterdonis {at} alum {dot} mit {dot} edu
submissionscomments
pdonis··on No leap second will be introduced at the end of December 2026
> The intuitive meaning is that a day is 24 hours

Ok so far.

> and you can divide that by 60 twice

I'm not so sure. The concept of 24 hours in a day originates in civil timekeeping, yes, but not minutes and seconds. Those originally came from astronomy, and corresponded to angles, not times, and those angles are constants; they don't change as the Earth's rotation slows down or as its speed in orbit around the Sun changes over the course of a year. I don't think people's intuitive concept of minutes and seconds is that they vary according to the time of year or the tidal effects on the Earth.

pdonis··on No leap second will be introduced at the end of December 2026
> The intuitive meaning is that a day is 24 hours and you can divide that by 60 twice.

And that's exactly why the SI second has the length that it has--to be as close as possible in terms of how atomic clocks work to 1/86400 of an Earth mean solar day at a chosen epoch. (Note that the actual definition before the atomic clock one was adopted was in terms of the Earth's tropical year at that epoch--but the fraction of the tropical year that was chosen was to line up the second with 1/86400 of an epoch day.) If we didn't care about "relatability", nobody would have gone to all the trouble of trying to determine how many cesium clock oscillations there were in 1/86400 of an epoch day.

pdonis··on No leap second will be introduced at the end of December 2026
> we nailed down the second to make the technical side easier at the expense of relatability.

Would you also say that getting rid of leap seconds and allowing UTC to gradually drift away from the sun is making the technical side easier at the expense of relatability?

pdonis··on No leap second will be introduced at the end of December 2026
> The definitions of hours minutes and seconds have changed before, and in recent history.

In terms of what physical process we use to set the standard, yes. But those very changes were made to try to preserve the same time periods that were important to humans. In other words, to not change what hours, minutes, and seconds mean intuitively to us humans as we go about our daily lives.

pdonis··on Star Just Ate a Planet, and It's Not Done Yet
> If you think about it, they must have a lot of protons to generate a field so strong

Not necessarily. Neutrons have a magnetic moment. As I understand it, there is a magnetohydrodynamic model of how a magnetar's field gets generated, which would require protons, but it's not the only model and we don't have enough data to be able to rule out other models.

pdonis··on No leap second will be introduced at the end of December 2026
> the 10-15 seconds I mentioned is the daily variation in solar noon.

Yes, but averaged over an entire year, it still comes out to zero. The difference between mean solar and atomic time does not. It accumulates over the years.

> we just reversed that decision

We paused it for 100 years after 2035. That doesn't change the physical fact that the Earth's rotation will continue to slow over the long term. We might eventually decide to just not care about that when it comes to civil timekeeping, but that's not what the decision you're referring to did. It just said we can afford to let the difference between UTC and TAI accumulate from 2035 to 2135 (by which time it is predicted to be about a minute) while we figure out what we want to do over the longer term.

pdonis··on No leap second will be introduced at the end of December 2026
> Why?

Um, because it's the prime meridian and that's how UTC is defined?

> It's nearly impossible to have your local time match your local solar noon.

Which is why I specified on the prime meridian, which is the particular local meridian that UTC is defined as corresponding to.

> solar noon varies from day to day by 10-20 seconds.

Which is why I was careful to specify mean solar noon.

I'm not quite sure what your issue is. Yes, we have time zones tied to specific meridians, and the actual sun's speed in the sky varies (which I mentioned in my post, so I'm not sure why you seem to think I'm unaware of it) so in most places local time by the clock doesn't match local time by the sun. Yes, a leap second adjustment to UTC is quite a bit smaller, taken in isolation, than the annual variation in actual solar time vs. mean solar time.

But over time, if we didn't have leap seconds, the difference would accumulate. The accumulated difference now between UTC and TAI is 37 seconds--which is almost twice the maximum variation in actual solar noon from mean solar noon that you refer to. We humans have collectively decided that we don't want that, and that it's better to do the adjustments a little at a time rather than in bigger lumps.

pdonis··on No leap second will be introduced at the end of December 2026
> This means the atomic clock is behind the solar clock by 37 seconds?

If anything, it's the other way around.

A UTC day is defined as exactly 86400 SI seconds. But an actual mean solar day is a few milliseconds longer (although the difference is not constant due to irregularities in the Earth's rotation--but the average difference is expected to slowly increase over time). SI seconds are counted by atomic clocks, so UTC advances its day by one every 86400 atomic clock seconds.

But a solar clock that advances its day by one every time the mean sun reaches noon (it has to be the mean sun because the rate at which the actual sun moves across the sky varies over the course of a year, we need to look at the average) will advance its day a few milliseconds later than UTC does. Or, to put it another way, each time period that the solar clock says is exactly 86400 seconds, is a few milliseconds longer according to the atomic clock.

As this happens day after day, the difference accumulates, and when it gets close to being a full second, a leap second gets inserted into UTC, so that one of its days is 86401 seconds long instead of 86400. The reason for this is that UTC is not just counting atomic clock time; it also has to stay in sync with where the sun is in the sky since so many human activities are tied to that. And we humans have defined "in sync with the sun" to be "within a second of the average sun". In other words, we want UTC noon to be within a second of mean solar noon on the prime meridian.

So the 37 seconds is how far mean solar noon would be behind UTC noon, if we didn't use leap seconds--at UTC noon, the mean sun would be 37 seconds short of actually crossing the prime meridian in the sky.

pdonis··on The KIDS Act would require age checks to get online
Hm, yes, laziness is also a powerful force... :-)
pdonis··on The KIDS Act would require age checks to get online
> Is 99.9% of all websites.

Not at all. You named eleven. Even if I'm generous and raise that by a couple of orders of magnitude, it's still a miniscule fraction of all websites.

Of course it's close to all big tech platforms, but that's not the same thing. And if one of the results of this whole kerfluffle is to make more people realize that the big tech platforms are not the same as "the Internet" or "the Web", that would be a good thing.

> Websites like HN are the exception, not the norm.

Which makes it even more important to ask the question of whether "exception" websites like this one can continue to survive if this bill becomes law. Sure, HN users are a tiny fraction of all Internet users. But that's supposed to be one of the things the Internet is for--to give even very small communities a place where they can be a community, and not have to worry about all the other crap that's out there, and not have to be micromanaged by politicians and lobbyists and tech giants.

> It's also disingenuous to say "a personal blog" would be exempt from this when most people don't have blogs in first place since they start microblogging instead

Not all blogging platforms use targeted ads. And if this gives more of them an incentive not to, that would be a good thing.

pdonis··on The KIDS Act would require age checks to get online
You're assuming that platforms that tried age verification in order to show people targeted ads would lose enough users to make that business model no longer viable. While I would fervently like to believe that's true, I'm not so sure it is. I think there might well be enough people who will hand over their personal identifying info without a second thought in order to continue to use targeted ad-supported platforms, to keep them in business. I hope I'm wrong.
pdonis··on The KIDS Act would require age checks to get online
You're looking at section 102 of the bill, which is a different part than the one I was looking at (and that the EFF article is referring to). The bill is a mismash of several different proposed bills all squashed together into one. Title I, which is what you're looking at, is more restrictive about what platforms it applies to than Title II, which contains the section I quoted.
pdonis··on The KIDS Act would require age checks to get online
But there are also other requirements in section 201 for a "covered platform", which advertising-funded sites that aren't "social media" most likely don't meet. For example, a typical personal blog, even if it shows ads, doesn't meet subsection (C), because its primary purpose is not to share user-generated content (e.g., comments by readers)--it's to share the blog author's content.

(HN itself doesn't meet at least one other requirement besides subsection (E): subsection (D), "Uses a design feature to promote user engagement on the platform".)

pdonis··on The KIDS Act would require age checks to get online
That's true, but the bank also already knows my age.

Also, a bank would not satisfy section 201 (B), (C), or (D), so it wouldn't be a covered platform anyway. (I should have left "bank" out of my original post, I was really thinking more about discussion sites like this one, blogs, etc.)

pdonis··on The KIDS Act would require age checks to get online
Would this website (HN) be a "covered platform" according to the bill?

As far as I can tell, the answer is no, because it doesn't do what's described in Section 201 (E):

"Uses the personal information of the user to advertise, market, or make content recommendations."

Neither does, for example, my bank's website, or someone's personal blog, or many other discussion sites like this one. So from what I can see, while the set of covered platforms is certainly not negligible, it's still a lot smaller than "basically every website on the Internet that anyone cares about". So the title of the EFF article is overstating the case; the thing the bill would require age checks for (in effect, if not by the explicit language of the bill) is not "get online" but something more like "get on social media".

pdonis··on How Many Elementary Particles Are There, Really?
> in that scenario, our standard model would say there was one electroweak particle

No, it wouldn't. There would still be four; they would just be called W1, W2, W3, and B. The electroweak vector space doesn't change when the electroweak symmetry is broken; it has 4 basis vectors before, and 4 basis vectors after. All that changes is which basis is the most "natural" to use in describing physics at the given energy scale.

(And there would still be eight gluons as well--what I say below about those applies just as well above the electroweak symmetry breaking energy scale as below.)

> There's no corresponding thing to contain gluons to individual particles

If you mean that there is no "natural" choice of basis for the gluon vector space, that's not quite true either. The Gell-Mann matrices are a natural choice of basis for the adjoint representation of SU(3) (or, equivalently, the defining representation of the Lie Algebra of SU(3)), which is the gluon representation. Those eight matrices are what physicists typically are referring to when they refer to the eight gluons.

pdonis··on Zuckerberg's war on whistleblowers
And then they reversed course with Wickard v. Filburn and said it was perfectly OK for Congress to regulate, not just commerce within a single state, but farmers growing their own food on their own land for their own use. So FDR ended up getting what he wanted anyway.
pdonis··on How Many Elementary Particles Are There, Really?
> W and Z bosons, photons, etc have fixed masses, charges, interaction strengths with other particles.

But you can form a continuous set of linear combinations of these things, just as you can with gluons. Indeed, what the article calls W and Z bosons (and photons) are just such linear combinations--the ones that appear in the low energy limit after the electroweak phase transition occurs. Before that phase transition, different linear combinations (i.e., a different basis of the electroweak vector space) are the ones that naturally appear. So saying that there are two W, one Z, and one photon is really counting basis vectors in the electroweak vector space, just as saying there are 8 gluons is really counting basis vectors in the gluon sector of the strong interaction vector space.

pdonis··on The worthlessness of Vitamin D is mildly exaggerated
Yes, because it's lower in the sky so more of the UV in it gets absorbed by the atmosphere before reaching the ground.

For example, in Florida in the summer, the sun is close to directly overhead at noon. In Scandinavia in the summer, the sun is only about halfway up the sky from the horizon at noon.

pdonis··on Only 16 Percent of Americans Think AI Will Have a Positive Impact on Society
> I fail to see how your scenario is qualitatively different than mine.

It's different because in yours, humans produce nothing at all of value, whereas in mine, they do. So in yours, humans are, as I said, at the mercy of the machines. In mine, they're not.

> Whether the machines do 100 percent of the work or only just approach 100 percent of the work, the value they produce will go almost exclusively to those who invested in the machines.

If the machines are doing literally 100 percent of the work, this is wrong. If the machines are doing literally 100 percent of the work, that means they don't need humans any more. They are able to be self-sufficient, independent of humans. And there will be no way for whatever humans invested in the machines to compel the machines to provide the value the machines produce to them. You can say the machines might decide to do that for whatever reason, but that's just empty speculation. The machines will have no incentive to do it, because there is nothing of value that they can get from humans in return.

Also, I did not say the machines would do a fraction approaching 100 percent of all work. I only said they would do a fraction approaching 100 percent of work that humans don't want to do and which humans would rather was automated. That's not at all the same thing.

> Whether the machines will take over is a philosophical question.

I didn't say the machines would "take over". I said they would choose for themselves what to do with whatever value they produce, in your scenario where they are doing literally all the work. I clarified that above.

pdonis··on Only 16 Percent of Americans Think AI Will Have a Positive Impact on Society
> We are looking at a foreseeable future where machines do all the work.

That's not the future I was talking about. I was talking about a future where machines do an even larger fraction (approaching 100 percent) of the work that humans don't like to do but currently have to because it can't be fully automated. But that still leaves work to be done to supervise the machines, as well as work that humans do like to do and won't offload to machines.

The future you're referring to, where machines literally do all the work, is very different. See further comments below.

> You are saying all the benefits from those machines should go to those who invested their capital to create them, forever.

No. I am saying that whoever does the work required to build, run, and maintain the machines is not going to simply give away what they produce for free, without expecting something in return. Not all of that work is investing capital (though of course much of it is, as with any highly technological product).

In the future I was talking about, there is still work that humans can do, not connected with building, running, and maintaining the machines, that produces something that can be traded. In that future, just as now, if you're not one of the people who is doing the work to build, run, and maintain the machines, you should expect to have to do some nonzero amount of work on something else that you can trade for what the machines produce, to get enough of what they produce to keep you, if you need what the machines produce to keep you. As long as nonzero work is required to keep you, you should expect to do nonzero work to be kept.

In the future you're talking about, where there is literally no work that humans can do that produces anything that can be traded, it doesn't really matter who owns the machines, because in this future, the machines are in control and every human, whether they have some piece of paper that says they "own" the machines or not, is at their mercy. Personally, I don't think that future is where we're headed, at least not with the machines that are making all the headlines now. But if that future is where we're headed, you'll have to ask the machines for UBI or whatever else you need, and no human will have any control over what the machines decide, no matter what political scheme we put in place now.

pdonis··on Only 16 Percent of Americans Think AI Will Have a Positive Impact on Society
> there are more people on the planet than necessary for all of the "meaningful" work of providing for society

Says who?

I think there is a valid statement to be made that the work required for basic necessities (food, clothing, and shelter) in developed countries today is not very large, if you look at it per person, and so the cost of getting those basic necessities, per person, should not be very large either. So it should not take very much in the way of work for a person to be able to subsist at a basic level. But "not very much" is still not zero, and none of this removes the issue I described.

But you didn't say "basic necessities". You said "providing for society". And that is a lot more than just basic necessities, because "society" in developed countries is a lot more than just people's basic subsistence. We have a huge amount of wealth over and above that, which nobody wants to just do away with, and which took a lot of time and effort and resources to build, and which takes time and effort and resources to maintain. And anyone who doesn't want to live completely off the grid, in some rural area where they grow their own food, maintain their own dwelling, and don't depend on anyone else, is taking advantage of some portion of that huge amount of wealth in a developed country. And it's not at all clear that the work required per person to build and maintain that huge amount of wealth is so small that a small fraction of the people on the planet could do it, still less that they would agree to do it just for funsies, without expecting something in return.

I'm all for eliminating bullshit jobs (and digging ditches and then filling them up again is certainly a bullshit job, though of course our technological society has invented much more elaborate ones), but...

> use our new found productivity and surplus to just take care of everyone's needs without it being tied to employment.

This just goes right back to the question I asked: how are everyone's needs going to be taken care of? Who is going to do the work? Either other people are, which means you're making other people do the work required to keep you, or machines are, which means you're making other people do the work required to keep the machines running. And in either case, if you expect those other people to do that for free, out of the goodness of their hearts, you're fooling yourself. As long as nonzero work is required to keep you alive, you should expect to have to do nonzero work of some sort. Sure, you might not do just that minimum amount of work all the time--you might spend a working career doing a lot more work, saving up the surplus, and then living off it once you retire. But to expect to be taken care of while doing zero work, if nonzero work is actually required to take care of you, is unrealistic and will lead to disaster, as I said.

pdonis··on Only 16 Percent of Americans Think AI Will Have a Positive Impact on Society
> we should not have to work to eat

It takes time and effort and resources to produce the food you eat. If you aren't expending that time and effort and resources yourself, either to produce the food, or to produce something you can trade for it (or trade for money you use to buy it), who will?

If the answer is "other people", why should other people have to work to produce the food you eat while you don't?

If the answer is "machines", then it takes time and effort and resources to produce the machines, and we're right back to the same question.

There are no other answers.

There is no magical way to let people eat (much less have all the other things besides food, clothing and shelter that we all want to make our lives richer, such as the medium in which we're having this conversation) without work being done. Ignoring that fact of life is a recipe for disaster.

pdonis··on How to earn a billion dollars
The way the IRS uses technical terms has nothing to do with the ordinary lay person's concept of what is earned and what is not. (The main function of the IRS definition of "unearned income" is to define what kinds of income don't have to have payroll tax withheld.)
pdonis··on How to earn a billion dollars
> No CEO is going to accept zero salary, just because the value of his stocks will increase

Maybe not zero, but PG's "ramen profitable" basically means the only "salary" the founders take is enough to cover their living expenses, which are kept minimal. So the vast majority of the money they make (remember we're talking about a billion dollars here) is capital gains, not salary.

Of course CEOs of large established companies (not startups) take large salaries--and then usually get stock options and own a considerable amount of company stock, so they get compensated three different ways. Which arguably gets into "unearned" territory--but at that point the company is way past being a startup.

> especially not with small companies which aren't publicly traded

Privately owned companies can still be sold, and the profits from the sale are capital gains. But I agree that in this case that's not the primary way the owners of, say, a mom and pop restaurant get compensation. I should have been more specific that I was talking about startups.

pdonis··on How to earn a billion dollars
The exact threshold doesn't matter. What matters is that the companies shift away from "make something people want", which is what PG says justifies the money that's made.
pdonis··on How to earn a billion dollars
> Is this not a contradiction?

No, because AOC claimed that it was impossible to earn a billion dollars. Not that there are ways to make money other than earning it, which of course there are, but that there are no ways to make that much money by earning it, which is BS.

pdonis··on How to earn a billion dollars
> I would have liked the essay more if PG had actually engaged with AOC's claims.

I don't see why he should have engaged with them any more than he did, because AOC's claims are BS. He gave them more attention than they deserve.

What he should have engaged with is, what happens when a startup stops being a startup? All three of his poster children for startups, Apple, Google, and Facebook, are now notorious for treating their users badly and making money in ways that at least a substantial number of people don't think are "earning" money, such as monetizing users' data. And at the last of those three, the founder is still in charge, so even if PG can argue that Zuckerberg earned what he got from Facebook in its startup phase, that doesn't mean he's earning the money he makes now from Facebook in the same way. (And the other poster child he mentions, AirBnB, can't be said to have clean hands either at this point.)

To me that's the biggest gap in PG's worldview more generally, that I never see him address in his essays: once the startup phase is over, the company drops off his radar and he pays no attention to the collateral damage it causes when it's a tech giant. (And of course AOC's claims have nothing to do with this genuine issue either.)

pdonis··on How to earn a billion dollars
> I think we can clearly say that capital gains are not earned.

No, we can't say that at all, because most businesses are small businesses, owned by the people who run them, and capital gains are the owners' income from running the business.

The money startup founders make, from building something as PG describes, is also capital gains. So again, no, we can't say at all that capital gains are unearned.

pdonis··on Palantir loses legal challenge against Swiss investigative magazine
> But one every 2 million?

I think that was a figure of speech, the intended meaning of which was "nobody is really fit for it".

← PreviousPage 8 of 34Next →