Hacker News: p0ckets

New comment by p0ckets in "What's happening inside the NIH and NSF"

p0ckets — Wed, 05 Feb 2025 06:20:06 +0000

The US pays much more for drugs than any other country. I guess one possibility is that no one has any leverage and big pharma is able to charge Americans more because they're richer. But the more popular theory is that countries can negotiate better prices than individuals can (yes, technically insurance companies can negotiate prices, but they seem unmotivated to drive any costs down). It seems the previous administration thought the government can negotiate lower prices: https://www.cms.gov/inflation-reduction-act-and-medicare/med....

New comment by p0ckets in "Take the pedals off the bike"

p0ckets — Wed, 15 Jan 2025 17:13:45 +0000

I always wondered why almost all the kids bikes sold in Canada have coaster brakes. I actually thought all kids bikes had coaster brakes until I saw some at Decathlon with hand brakes (turns out it's a French company).

New comment by p0ckets in "The Skyline algorithm for packing 2D rectangles"

p0ckets — Tue, 19 Nov 2024 17:41:11 +0000

Not mine. I just happened to hear about it from a colleague recently.

New comment by p0ckets in "The Skyline algorithm for packing 2D rectangles"

p0ckets — Tue, 19 Nov 2024 00:55:57 +0000

An even more recent one from Google: https://github.com/google/minimalloc

New comment by p0ckets in "Is a 'slow' swimming pool impeding world records?"

p0ckets — Wed, 31 Jul 2024 19:32:20 +0000

"minimum standard of 2 meters that was still in place when Paris 2024 plans were approved; but below the new World Aquatics minimum of 2.5 meters."

Although the recommendation has been 3 meters for a while.

New comment by p0ckets in "How do you accidentally run for President of Iceland?"

p0ckets — Mon, 29 Apr 2024 19:20:54 +0000

Zelenskyy seems to have risen to the occasion.

New comment by p0ckets in "It Is Somebody's Moral Imperative to Leak "Coyote vs. Acme" to the World"

p0ckets — Tue, 13 Feb 2024 04:07:35 +0000

What I mean is this movie is so bad it will ruin it for the next "Coyote" movie, so they don't want anyone to see this movie. Maybe the only compromise that works is to put it in a time capsule and not open it for 100 years.

New comment by p0ckets in "It Is Somebody's Moral Imperative to Leak "Coyote vs. Acme" to the World"

p0ckets — Tue, 13 Feb 2024 03:49:02 +0000

There is some value in not having a "Coyote vs. Acme" film released this year (either by Netflix or the Library of Congress) because they could try to make a profitable version soon but people may not give a new version a fair chance if a really bad version was released recently.

They presumably value that option more than whatever Netflix et al. offered.https://www.theverge.com/2024/2/9/24067496/coyote-vs-acme-am...

New comment by p0ckets in "Bank financing and bad urban planning make the retail apocalypse worse"

p0ckets — Thu, 02 Nov 2023 03:58:59 +0000

I think your first proposal addresses the root cause.

But the second proposal - forcing municipalities to write options with unlimited potential losses - is silly. The whole point of the zoning authority is to say "it'll be best for the neighborhood in the long run for there to be some retail shops here", and it makes sense whether you're talking about a lot or the first floor of a building. If the development isn't profitable with the ground-floor retail requirement, then just don't build it.

Some sort of option might work, but it would have to be something like: the municipality can choose to rent the space for a fixed price if it's vacant for long enough. This encourages the landlord to rent it for some higher price.

New comment by p0ckets in "FCC refuses to scrap rule requiring ISPs to list every monthly fee"

p0ckets — Wed, 30 Aug 2023 20:29:04 +0000

According to the article: "all charges that providers impose at their discretion, i.e., charges not mandated by a government." So they wouldn't need to include your school excise tax.

If they charge different fees depending on address (again, not taxes) then it seems like the intended effect to provide the actual price the consumer will pay.

New comment by p0ckets in "Automakers are starting to admit that drivers hate touchscreens"

p0ckets — Thu, 27 Apr 2023 02:04:33 +0000

That's basically what Audi has (at least in some older models): it's a scroll wheel that also works as a d-pad. It works pretty well with Android Auto (no experience with Apple Car Play). Mazda also has a similar-looking interface.

But it seems those two companies are moving towards touchscreens due to market demand.

New comment by p0ckets in "South Korea to give $490 allowance to reclusive youths to help them leave house"

p0ckets — Thu, 13 Apr 2023 17:38:24 +0000

Singapore has this: https://en.wikipedia.org/wiki/Social_Development_Network, although I'm not sure how successful hikikomori will be at these events...

New comment by p0ckets in "Hertzbleed Attack"

p0ckets — Tue, 14 Jun 2022 19:27:53 +0000

And for this particular attack, constant time isn't even enough! You would need either constant power, or limit the frequency when running secure code (which again reduces performance).

New comment by p0ckets in "Hertzbleed Attack"

p0ckets — Tue, 14 Jun 2022 19:21:55 +0000

Until consumers demand this as a requirement, it won't happen. Almost everyone would rather have a compiler/language/OS/ISA/CPU that's finishes faster some of the time, rather than one that finishes at the same time all the time. It would just appear (especially in benchmarks) to be slower for no apparent benefit.

Maybe we can introduce a new set of instructions that are guaranteed to be constant time, but good luck convincing the compiler/language/OS to use these slower instructions even if just for the code that is important for security.

New comment by p0ckets in "Namecheap: Russia Service Termination"

p0ckets — Tue, 01 Mar 2022 01:22:14 +0000

I shall invoke Godwin's law, and compare this to IBM supplying Nazi Germany with computers: https://en.wikipedia.org/wiki/IBM_and_World_War_II

Do you honestly believe that the right thing for IBM to do was honor German contracts after the USA and Germany declared war on one another? If so, would German bombardment of American cities change your mind?

New comment by p0ckets in "The Minimum Viable Testing Process for Evaluating Startup Ideas"

p0ckets — Fri, 17 Sep 2021 17:23:48 +0000

Paying someone to plan a trip for you is definitely not a product, so not a MVP. Running a course is closer to a full-fledged product (more than a MVP) for a company that sells courses to students, but Maven is a platform for creating courses...

By the definitions laid out in the article waitlists and crowdfunding campaigns would be MVTs to address marketing or "building something people don’t want" risk.

New comment by p0ckets in "All public GitHub code was used in training Copilot"

p0ckets — Thu, 08 Jul 2021 15:05:36 +0000

I think to actually poison the well, we should add code to existing repos with dead code clearly labelled as "the way that things shouldn't be done" that are wrong in subtle ways. So every time we fix a security issue, we keep the version with the bug with some comments indicating what's wrong with it. Of course, this only works until the AI is trained to weigh the code based on how often the code is called.

New comment by p0ckets in "The True Meaning of Technical Debt (2020)"

p0ckets — Tue, 13 Apr 2021 14:09:50 +0000

Like a kind of inflation that devalues both code written and associated technical debt.

New comment by p0ckets in "Gumroad raises $5M from Crowd SAFE in first 24 hours"

p0ckets — Tue, 16 Mar 2021 14:37:16 +0000

Do we know for a fact that the investors sold the equity back to the founder and not the company (like a stock buyback)? The investors just wanted to get out, so tracking down former employees, deciding how much equity each one deserves, and selling them equity probably isn't what they want to spend time doing. So the only real options would be selling to the company, or to the founder.

New comment by p0ckets in "Are Humans Intelligent? An AI Op-Ed"

p0ckets — Wed, 05 Aug 2020 18:17:43 +0000

But only the best examples of GPT-3 output will be worth putting on the web, so even without any code improvements in future GPT-x models you can still expect an improvement in quality. A writer (human or AI) should be able to improve by writing a million articles and then be told which ones are good (i.e. worth publishing).