HNHacker News
TopNewBestAskShowJobs

knzhou

5,522 karma · joined September 2, 2019

I'm a physics graduate student. You can also find me on:

Physics StackExchange: https://physics.stackexchange.com/users/83398/knzhou

Personal website: https://knzhou.github.io

submissionscomments
knzhou··on Evidence for Anyons, a Third Kingdom of Particles
Yeah. In this case it's fine, but in general it can lead to confusion.

For some weird reason, condensed matter physicists keep giving their quasiparticles the same names particle physicists give their particle particles, even though the mathematics are quite different. If you hear someone proclaiming the discovery of Majorana fermions, axions, or monopoles, it's essentially guaranteed to be a quasiparticle in a lump of metal or something. Sometimes even the articles themselves get confused about this.

knzhou··on Evidence for Anyons, a Third Kingdom of Particles
No, Quanta is fine -- you're making a subtle error, which is in fact common even in graduate textbooks.

There are two distinct things going on here: formally "swapping" the arguments in the wavefunction, and physically moving the particles around. A priori, there is no reason the phase shifts from these two actions should be related.

You're focusing on the swap, but it's not interesting: your argument that the swap phase squares to 1 applies to both 2D and 3D equally well. When people talk about anyons, they're talking about the phase from moving the particles. That has more options in 2D because the paths one can take when moving particles around can have a greater range of topologies (https://en.wikipedia.org/wiki/Braid_group).

The reason that textbooks mix these things up is because in relativistic quantum field theories these two phases are always the same, by virtue of the spin-statistics theorem. But the anyons of condensed matter are nonrelativistic.

knzhou··on A non-moral dilemma about cheating, but it brings up some points (2019)
I was lucky to spend my childhood and college years in peer groups where cheating was incomprehensible. Nobody would cheat in a math competition, for instance. That would just be a loud, lame declaration of stupidity.

When I went to the other side and started having to enforce the rules, I realized that while the top decile could be trusted, everybody else was suspect, and enforcement of the rules was incredibly hard. What do you do when you grade a string of tests containing the exact same silly mistakes? What about a group of students that suddenly jumps to high performance on one exam and then immediately crashes back down? Remote proctoring is even harder. I once saw a student fabricate an entire school with the aid of his parents, complete with website, address, and fake teacher information, so he and his parents could do the exam together. How are you supposed to vet hundreds or thousands of people like this?

I really hope that I get a faculty position where I can just teach low-pressure graduate-level seminars, and never have to learn the answer. Mass education is damn hard.

knzhou··on A 2020 Vision of Linear Algebra
Gilbert Strang's linear algebra course blew my mind back in high school, and I still use insights from it every day. Strang has a particular lecturing style where he approaches every topic several times, often beginning many lectures before the main treatment. At first I thought it was a bit confusing, but later I realized it helped build fluency, just like a language class.

I'm really thankful to MIT OCW for putting his lectures out for free -- in fact, I think I'll go donate to them now.

knzhou··on If you overlay all atomic spectra, you get a Planck distribution
Neat! Still, I wouldn't say the Planck distribution is noticeably better. The reason the other 4 distributions appear to have the peak too low is because they're doing a tradeoff: they do that to better fit the values of the left hand side of the graph.

The Planck fit gets the peak height right, but mainly because it has no chance of fitting the left hand side at all (since it increases much more slowly than the other distributions), so it doesn't even try.

This is why I said it would be better to have actual statistical tests -- we shouldn't have to have this kind of qualitative discussion when it's already a solved problem.

knzhou··on If you overlay all atomic spectra, you get a Planck distribution
I've seen this blow up in several places now, but there seriously is nothing to this paper.

Essentially all they're pointing out is that the set of atomic transition energies (1) is positive, (2) has a smooth distribution, (3) goes to zero at zero and infinity, (4) has a maximum somewhere in between, (5) is skewed right. All of these things are completely mundane and well-understood, and not at all unique to the "Planck distribution". Statisticians probably know of tens of other distributions with these properties, which would fit their curve about as well.

Their claim is like saying that any function that goes from -1 to 1 smoothly must be a logistic function, or any function that goes to 0 at infinity but slowly must be a power law. That's not a paper, that's a hunch. If the researchers wanted to be serious, they could have run a statistical test to quantify how well the data fit the Planck distribution (just like tests of normality are routinely done in statistics). But they didn't, and the reason probably is because the test would fail.

knzhou··on Analysis of the Imperial College Epidemiological Model
These options are so soft that they really are equivalent to doing nothing. Sick people have always been told to stay home; trips abroad and on ships might cause visible clusters early on, but they're a drop in the bucket later when the disease is already at home.

The real issue is that a strong response causes people to take things seriously, and a weak response causes them to not take it seriously. In other words, just like the model, we have a situation where predicted results inform the actions taken, which affect popular opinion, which in turn change the predicted results.

Yes, the optimal outcome is to somehow have the government do nothing, thus restricting nobody's freedoms, but simultaneously have society take it so seriously that they eradicate the disease themselves in two weeks. But that is not a genuine policy option.

knzhou··on Analysis of the Imperial College Epidemiological Model
We could argue about code smells all day, but ignoring that, I think there's only one essential point made in the post. It claims that without a lockdown, things would be just fine, because people would automatically isolate themselves on their own. And therefore the model is useless.

This is a completely unfair criticism. When the relevant decisions were being made two months ago, a huge number of influential people in the UK were seriously proposing to do nothing to get herd immunity. This model, which intentionally depicts what would happen if people indeed did nothing, was a big part of why people changed their minds.

You can't say the model is wrong because people would take COVID seriously anyway, when the model is half the reason people take it seriously in the first place. We're somehow forgetting history that's only months old now.

knzhou··on Finding new physics will require a new particle collider
These arguments were indeed used 30 years ago, but shortly after the USSR fell apart, the American Superconducting Supercollider was defunded midway through construction. The US was satisfied to become 2nd to Europe in this field then, I don't see why the US wouldn't be equally satisfied to be 4th to Europe, Japan, and China a little while later.
knzhou··on Finding new physics will require a new particle collider
Perhaps counterintuitively, it makes little difference. At the energies we're considering, the mass-energy of the particles being collided is completely negligible. Switching to colliding protons and antiprotons does change the details of what you get out, but not the energy scales.
knzhou··on If Planet Nine Is a Tiny Black Hole, This Is How to Find It
When the original paper about planet 9 being a black hole came out a year ago, everybody thought it was a cool-sounding but just completely wild idea -- both physicists and pop science articles. Not impossible by any means, but not within the top 250 new physics ideas in plausibility. Then the discussion died down. I figured it would end up being one of those forgotten but cool gems.

Then the most authoritative string theorist in the world, who hasn't worked on anything relating to experiments in decades, suddenly writes a very short, simple, single-author paper to use this idea to promote Yuri Milner and Mark Zuckerberg's space probes. Now it's the conjunction of two crazy ideas, and it's really taking off. I just don't know why this is happening, or what to make of it.

knzhou··on Finding new physics will require a new particle collider
Nima-Arkani Hamed likes to point out that particle colliders aren't getting more expensive at all, in relative terms. The price tags are going up for just the same reason that you can't buy a loaf of bread for a nickel anymore, but as a fraction of GDP, the next big collider has always been about 0.01%.

In fact, if there's any identifiable trend at all, it's that this fraction has been falling. The LHC, for instance, was slapped together after the last would-be supercollider was defunded, and reuses the LEP tunnel (built in the 1980s). If you're wondering why we took so long to find the Higgs boson, it's because the field has already been shrinking under the pressure of declining funding for 40 years.

knzhou··on Finding new physics will require a new particle collider
The actual timeline for this is that when the first critiques of LIGO's analysis came out, there was a good reply from LIGO and a productive scientific discussion, at the end of which Sabine indicated she was satisfied. That was months before. That was good.

Then she came out with the linked post, which pretends LIGO had no reply to the substantive issues even though she already had blogged positively about their reply earlier, and disingenuously throws in a few complete non-issues just to make a gish-gallop. That's not science popularization, that's spreading FUD for the sake of self-promotion. Any field can be "taken down" with enough dishonesty.

knzhou··on We are not mostly empty space
But visible light can't get through us at all. Does that prove that we aren't empty space? The whole point is that this criterion is incredibly ambiguous.
knzhou··on There is one, and only one strain of SARS-CoV-2
That’s not a useful comparison though. You want to compare cases now to cases when lockdown or similar measures were imposed. Otherwise a country that wastes 2 months doing nothing and then takes action far too late just gets put in the “lockdown” bin.
knzhou··on SARS-CoV-2 was already spreading in France in late December 2019
No, that's just how flu season works. The numbers in the US are also higher this year, and it normally ranges within a factor of ~4 from year to year depending on the dominant strains in circulation. [0] It's completely implausible that there were a million cases of COVID-19 in December.

0: https://www.cdc.gov/flu/about/burden/past-seasons.html

knzhou··on Why it's great to write blog posts
I recently argued [0] why this is a great thing in academia especially. Getting cutting-edge information out of papers alone is extremely difficult. Asking experts is ideal, but not everybody can do that. Lecture notes, blog posts, StackExchange answers, and the like all fill a critical intermediate role.

0: https://academia.stackexchange.com/questions/148261/do-you-k...

knzhou··on YouTube has deleted the account of David Icke
> So I gather that he's espousing a view that states that quarantines and lockdowns are terrible and ineffective, and encouraging people to ignore them and freely associate.

> I understand that you believe the expression of that view could lead to many people deciding to ignore whatever orders or ordinances that have been imposed upon them.

That is not at all a representative summary of his views, or mine. I see hundreds of people arguing against lockdowns every day, on sites all across the internet, and it's just fine, and nobody is being censored for it. It's so mainstream that even the NYT is publishing opinion pieces about it, e.g. when comparing with Sweden's response.

The actual problem with David Icke is that he espouses conspiracy theories, of the form "the lizard men, Bill Gates, the aliens, and the liberals teamed up to create 5G to wipe out the human race". Stuff like this leads to serious consequences, such as the Congressional baseball shooting, the Pizzagate shooting, the Youtube headquarters shooting, and the armed takeover of the Michigan statehouse.

knzhou··on YouTube has deleted the account of David Icke
No, because at the time the statement was made, most of the known cases were directly linked to a single wildlife market. (Also keep in mind that this announcement is within 2 weeks of discovering the virus, sequencing its genome, inventing the first ever tests for it, and deploying those tests!) Luckily, enough evidence came in that they were sure a few days later.
knzhou··on YouTube has deleted the account of David Icke
Nobody is talking about banning people like Didier Raoult. If you can't tell the difference between a professor and a guy who gives spittle-flecked rants about the lizard people, this discussion will not be productive. For example, we all know he did a terrible job with the hydroxychloroquine study, but of course he retains his position.
knzhou··on YouTube has deleted the account of David Icke
> now comes the classic counter-argument: that the evils of disinformation possess a virality that makes them spread far more easily than the truth, being that truth is nuanced and difficult to acquire whereas bad information is seductively simple

This is such a knockdown argument that I can't believe you just refused to address it!

The mirror image of David Icke is the genuine experts that are trying to get good, nuanced information out there, often fresh out of research papers. Most of these people have single digit reach, i.e. so little that they could spend their whole lives getting less reach that Icke gets in 5 minutes.

The asymmetry in potency between good and bad information is far higher today than at any time in the past, and I applaud Youtube for doing a tiny bit to stop the bleeding.

knzhou··on YouTube has deleted the account of David Icke
What forums are you talking about? Radio, mail, TV, and newspapers still have audiences of hundreds of millions. Internet forums are in decline, but those have always had completely negligible public reach (as in, down by more than 2 orders of magnitude) compared to the others.
knzhou··on YouTube has deleted the account of David Icke
I feel like we're going in circles here. Even if every single site and app coordinated to kick off David Icke, he would end up, at worst, as silenced as the average American in the year 2000. It just doesn't seem all that bad to me.
knzhou··on YouTube has deleted the account of David Icke
Interesting, do you have further reading on this to recommend?
knzhou··on YouTube has deleted the account of David Icke
Because, as I said a couple comments ago, the degree of one-to-many communication available to people has been continuously skyrocketing for decades, and currently stands at levels far beyond any seen in history. Even the heaviest possible strawman-censorship one could imagine on Youtube would only erase a small part of this rise. That's because the internet has allowed one-to-many communication to massively scale up, while almost by definition one-to-one communication can't.
knzhou··on YouTube has deleted the account of David Icke
Maybe legalistically it is, but my point is that it's like the NYT and radio and print in the sense that these are all routes for one-to-many communication. Restricting one route of one-to-many communication isn't the end of the world. (Meanwhile, AT&T facilitated one-to-one communication. It would indeed be horrifying to censor that -- which is why precisely nobody in this giant comments section is advocating for it, nor is it going to happen.)
knzhou··on YouTube has deleted the account of David Icke
So you're arguing that once the public knows the truth... then the public will know the truth? Does sound like a rather weak claim, doesn't it?
knzhou··on YouTube has deleted the account of David Icke
Who said that? The range of bullshit information you can find at a moment's whim remains extremely high, in fact higher than it ever has been in history.
knzhou··on YouTube has deleted the account of David Icke
This slope isn't nearly as slippery as you're saying. If YouTube began issuing a million times more bans, it might approach the quality of an average newspaper in terms of accuracy. Would we be living in a dystopia at that point? No, we'd be living just like we did in the year 2000, when YouTube wasn't around but newspapers were. Just like a nutjob in 2000, Icke still has print, radio, TV, going outside with a megaphone, and going door to door with pamphlets to promote his message. In fact, he's even luckier, because he still has access to the entire rest of the internet.

Indeed, the fact that we're here discussing Icke's massive popular impact is proof that in 2020, there is far more free speech than there was in 2000, when he was mostly ignored. We are at historically high levels of free speech. Freaking out over this is like thinking we've gone back to the Bronze Age every time the stock market goes down 1%.

knzhou··on YouTube has deleted the account of David Icke
There really is no need for hysteria here. If YouTube began issuing a million times more bans, it might approach the quality of an average newspaper in terms of accuracy. Would we be living in a dystopia at that point? No, we'd be living just like we did in the year 2000, when YouTube wasn't around but newspapers were. Just like a nutjob in 2000, Icke still has print, radio, TV, going outside with a megaphone, and going door to door with pamphlets to promote his message. In fact, he's even luckier, because he still has access to the entire rest of the internet.

But wait, you say: in the year 2000, Icke wouldn't have been able to get a following of millions on print or radio! They wouldn't have let him on! Indeed, so what this actually proves is that in 2020, there is far more free speech than there was in 2000. We are at historically high levels of free speech. Freaking out over this is like thinking we've gone back to the Bronze Age every time the stock market goes down 1%.

← PreviousPage 4 of 27Next →