Hacker News: Hussell

New comment by Hussell in "Copilot edited an ad into my PR"

Hussell — Mon, 30 Mar 2026 19:12:10 +0000

Decisions are made by people in the group, not by a notional single being "the corporation". It's individual people making decisions about whether to go for short-term profit or long-term sustainability. Hold them accountable, don't shift the blame onto a nonexistent entity.

New comment by Hussell in ""Over 1.5 million GitHub PRs have had ads injected into them by Copilot""

Hussell — Mon, 30 Mar 2026 15:48:57 +0000

"We" here likely refers to Tim and his current coworkers who were present to see this, not every current and future employee of Microsoft / Github. Try not to think of any organization or institution as a person, but as lots of individual people, constantly joining and leaving the group.

New comment by Hussell in "210 IQ Is Not Enough"

Hussell — Thu, 20 Nov 2025 16:13:56 +0000

The maximum IQ score anyone can get depends on the total number of people who have taken IQ tests so far. Even if every single person alive today took an IQ test (which is absurd in itself), the maximum IQ achievable would be between 190-197. In practice, I'd guess the maximum is somewhere between 170 and 185 (millions to tens of millions of IQ test results which were recorded).

Even then, you need special tests to distinguish between anyone with IQ higher than about 160 - all those people get the same (perfect) score on regular IQ tests.

So: claiming to have an IQ of 276? Bullshit. The guy whose parents claimed he scored 210 on an IQ test? Also bullshit. To get 210, there would have to have been ~500 billion IQ test results recorded.

New comment by Hussell in "Statistical Formulas for Programmers (2013)"

Hussell — Thu, 13 Mar 2025 16:27:51 +0000

The statisticians have a bunch of tricks to transform the formulas into more-easily computable forms, e.g. calculate both the average and the standard deviation in a single pass through the data instead of one pass to calculate the average and a second to calculate the standard deviation. Converting the math in here to efficient code isn't very easy.

New comment by Hussell in "Orb weaver spider glue properties evolve faster than their glue genes"

Hussell — Wed, 19 Apr 2023 18:08:35 +0000

The two species share a lot of glue genes, and the properties of their silks differ due to other genes affecting things like the ratios of glue proteins in the silk. Those other genes evolved faster than the glue genes themselves.

The title is technically correct, though misleading.

New comment by Hussell in "Orb weaver spider glue properties evolve faster than their glue genes"

Hussell — Wed, 19 Apr 2023 15:37:58 +0000

Nowhere in the paper does it say that individual spiders are varying their silk composition in response to conditions. The selective expression of proteins is a difference between the two species studied, not between individuals within one species.

The title is technically correct: the two species share a lot of glue genes, and the properties of their silks differ due to other genes affecting things like the ratios of glue proteins in the silk, and those genes evolved faster than the glue genes themselves.

New comment by Hussell in "Orb weaver spider glue properties evolve faster than their glue genes"

Hussell — Wed, 19 Apr 2023 14:35:06 +0000

The spiders haven't even been shown to change their silk in response to conditions. It's just two species of spiders each with their own silk recipe.

New comment by Hussell in "Orb weaver spider glue properties evolve faster than their glue genes"

Hussell — Wed, 19 Apr 2023 14:31:08 +0000

Since I happen to know these species: the spider pictured at the top of the article is an Argiope bruennichi (Wasp Spider), a European species which closely resembles A. trifasciata (Banded Garden Spider, one of the two species in the study).

New comment by Hussell in "A curated list of falsehoods programmers believe"

Hussell — Thu, 17 Mar 2022 16:00:36 +0000

For those who haven't seen this before, the ones you should look at first are:

Falsehoods about Names https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-...

Falsehoods about Time https://infiniteundo.com/post/25326999628/falsehoods-program...

Falsehoods about Addresses https://www.mjt.me.uk/posts/falsehoods-programmers-believe-a...

New comment by Hussell in "A curated list of falsehoods programmers believe"

Hussell — Thu, 17 Mar 2022 15:08:00 +0000

Is there an easy way to display a diff which shows all the new links since the last time this was on HN? (Sept. 8th, 2020)

Partial list:

Falsehoods about Airline Seat Maps https://duffel.com/blog/falsehoods-about-seat-maps

Falsehoods about Biometrics https://shkspr.mobi/blog/2021/01/falsehoods-programmers-beli...

Falsehoods about Plain Text https://jeremyhussell.blogspot.com/2017/11/falsehoods-progra...

New comment by Hussell in "YouTube is trying to reward “quality” content"

Hussell — Fri, 12 Apr 2019 19:19:17 +0000

400 hours/minute is 24,000 minutes of video per minute, or 241,920,000 minutes of video per week. Assuming a human can review 8 hours of video per day, 5 days a week (too high, but not by much), and that YouTube engineers can create an algorithm that automatically marks 95% of video (probably low, 95% correct labelling is easy to get; 99% is usually necessary to outdo humans), then you would need 5,040 humans.

Automatically labelling 97.5% would halve that. If employees can only review 6 hours of video per work-day, then it would increase by a third. Both, you'd need 3,360 humans.

New comment by Hussell in "Millitext – A subpixel text encoding font"

Hussell — Tue, 18 Dec 2018 00:18:59 +0000

Here's another subpixel font called The Flea's Knees: http://www.typophile.com/node/61920

It's 3 pixels tall (not including ascenders and descenders, which are another two pixels each), and uses the full range of colors available instead of just black, white, and the 6 fully saturated colors. I think it's much more legible.

New comment by Hussell in "How Writers Map Their Imaginary Worlds"

Hussell — Mon, 05 Nov 2018 18:47:52 +0000

> Real rivers go somewhere. They don't just stop in the middle of nowhere.

Yeah? Have a look at the Morghab River in southern Turkmenistan. It flows from some mountains in Afghanistan out into a desert, where it splits up into a delta and just sort of... stops. The satellite view on Google Maps is striking. (There are several canals connecting to the delta now, but historically the river stopped in the middle of nowhere.)

Rules of thumb like this aren't laws, just patterns which have exceptions. (That said, yes, imaginary maps are sometimes ridiculous.)

New comment by Hussell in "Was there a civilization on Earth before humans?"

Hussell — Sat, 14 Apr 2018 17:52:39 +0000

Basically: because it has a lot of characteristics in common with all the other, better preserved, asteroid craters we know of. So our hypothetical advanced civilization would have to have deliberately mimicked all the little details of an asteroid impact, not just created a big explosion using some other method. That's possible, of course, and if they did it well enough we'd never know. But it seems much more plausible that it was an asteroid impact.

New comment by Hussell in "Lojban"

Hussell — Tue, 16 Jan 2018 18:42:33 +0000

Eliminating syntactic (not semantic) ambiguity is indeed a noble goal, but I was kind of horrified when I first realized how it's done in Lojban. Practically every type of phrase has both its start and its end marked by a unique word or class of words, with the terminator usually optional in any context where dropping it would not result in any ambiguity. Knowing exactly when dropping a terminator would result in ambiguity is pretty difficult, and in practice is rare enough that seeing a terminator is something of a surprise and usually requires a trip to look it up in the dictionary.

It seems to me that a grammar based on strictly right (or left) branching (that's prefix or postfix notation, or Polish or reverse Polish notation, for the programmers and mathematicians out there) would eliminate the need for all these optional terminators (which are effectively optional parentheses to clarify the order of operations).

New comment by Hussell in "Date/Time Inputs Enabled on Firefox Nightly"

Hussell — Mon, 23 Oct 2017 15:27:43 +0000

That's a problem we already deal with by having labels for the date inputs which explain the expected format. This is especially necessary in Canada, where until recently there was no recommended format, and the government recommendation isn't even well known yet much less universally used. So, depending on what part of the country you live in, you'll expect a different date format, even though you're technically using the same locale.

So, to do what you want, I'd have to make different versions of all the supporting documentation for each possible date format, and then get each version translated.

Also, displaying the dates the user just entered in a different format is a usability problem. (Users will complain. A lot.) So, if I did what you want, I'd likely be asked to display all dates, everywhere on the site, in the format the browser happened to be compiled with. This is possible for dates in the database, but not so much for dates in static content like news releases.

Just to be clear: I have no objection to making it easy for a website to have date input widgets which use the user's locale. I just want it to be equally easy for a website to have date input widgets in a specific locale. For example, how am I to make a website comparing date input widgets for different locales?

New comment by Hussell in "Date/Time Inputs Enabled on Firefox Nightly"

Hussell — Mon, 23 Oct 2017 14:59:55 +0000

I want the content, number formats, date formats, map labels, etc. on my website to all use the same locale. Date input widgets are no exception.

I can't afford to give full support to all locales (translation costs are ridiculous), so detecting and supporting the user's locale seems to be pointless unless they happen to be using one of the limited set I can give full support to (en-CA and fr-CA in my case). It seems like a far better idea to allow the user of my website to choose one of the supported locales, and see everything in the same consistent format than to have a few input widgets on each page decide to configure themselves based on the locale the browser was compiled with.

Edit: also, consider testing. Am I supposed to download a bunch of variants of Firefox to test my locale support?

New comment by Hussell in "Silently Corrupting an Eclipse Workspace on Windows"

Hussell — Thu, 23 Mar 2017 13:07:59 +0000

Sounds like you're one of those people who use Linux.

New comment by Hussell in "Scrolling on the web: A primer"

Hussell — Thu, 16 Mar 2017 13:50:00 +0000

Here's an example of a 2000s era page with a table about midway down with its own scrolling: http://www.ibacanada.com/site.jsp?siteID=ON001

Does this not work on some modern browsers out there?

New comment by Hussell in "Cracking Minesweeper with Z3 SMT Solver"

Hussell — Sun, 05 Mar 2017 20:20:21 +0000

There's a minesweeper solver which works out all the probabilities of uncertain squares. http://mrgris.com/projects/minesweepr/demo/player/

The creator wrote an explanation here: http://mrgris.com/projects/minesweepr/