Recent submissions

Comment · 9 days ago
Ok, but Switzerland is not a part of the EU. They participate only by treaty. So there has to be more to this story (legally-speaking)
Comment · 11 days ago
> That doesn't work if you are writing a library where users of the library are able to implement their own event types.

-or- in places where you simply want to maintain a compilation firewall, for compile times, or encapsulation... i.e. you don't want to update the "visitor" in a pile of unrelated files each time you come up with a new Event.

Comment · 11 days ago
The company behind Google+ is not good at product rollouts? Impossible!
Comment · 14 days ago
Can Junie design a less-readable web page?
Comment · 20 days ago
> using a passkey in a flexible password manager (Bitwarden, keepass, 1Password, etc) really does continue to give the user control.

Sure, but if you read the article you would know that deal can (and will undoubtedly) be altered. The passkey issuer can demand that the passkey be hardware-backed, and I've already run into passkeys that cannot be stored in bitwarden out in the wild (i.e. they REQUIRE that I use a TPM-backed store in my computer). Eventually it will be a particular TPM-backed store running on an attested platform. Now that's not generally the case today, but make no mistake, that IS the ultimate goal... You will only access valuable services from devices that you fundamentally have no control over (e.g. locked down OS's with attestation before you can access X,Y,Z), so that you can't steal from netflix, skip the sponsor segments on youtube, block ads in your web browser, etc. etc. etc. If you want a preview of how this might all work from the technical side, just look into how TEE works for large-scale GPU deployments right now.

Comment · 2 months ago
meh, every single license key ever issued is stored in the binary.... also you can just compile your own copy from scratch.

-or- just pay the author the $7 or whatevs if you use it.

Comment · 3 months ago
That's the absolute best possible outcome... The more severe risk is where a neighborhood child gets severely injured and then the liability policy/coverage you were relying on as a homeowner is invalidated because you failed to disclose the trampoline (which is one of the things they typically ask for). Now you are stuck with legal + medical bills which will likely bankrupt you and/or cost you that house.

Same with pitbulls. Like "neat you hid it from your insurance company, boy you sure showed them!" They are going to be so super sad to cash all of your premium checks for years and THEN pay out $0 on the liability claim when fido rips a kids face off.

Comment · 3 months ago
> But during covid archive.org decided to just remove the limit and lend unlimited copies concurrently

IIRC, this was 100% it. Lending one digital version of one physical asset was likely already a violation copyright. Lending UNLIMITED digital versions of one physical copy was DEFINITELY a blatant violation of copyright.

Comment · 3 months ago
> I loathe their bug-ridden UI.

So weird that the same exact people telling you that programming careers are now obsolete are the same group who haven't been able to fix screen flickering bugs for like a year...

Comment · 3 months ago
searching for .gov reveals 0 matches... doubt
Comment · 4 months ago
>very loudly stating the foreign labs have been distilling their models

Help! Someone else is blatantly ripping off my plagiarism machine!

Comment · 4 months ago
> Fail cleanly.

This is the same exact industry that gives you paid usage limits as a unit-less percentage bar then gaslights customers every time the algorithm running that percentage bar changes or they lobotomize an existing model with increased quantization to squeeze a few more dollars out of existing hardware.

"Failing cleanly" might make their moated hype-machine look bad pre-IPO, so they certainly aren't going to do that voluntarily.

Comment · 1 year ago
Yes, but they are suuuuper safe. /s

So far I have mixed impressions, but they do indeed seem noticeably weaker than comparably-sized Qwen3 / GLM4.5 models. Part of the reason may be that the oai models do appear to be much more lobotomized than their Chinese counterparts (which are surprisingly uncensored). There's research showing that "aligning" a model makes it dumber.

Comment · 1 year ago
That is fine as long as the input / output is always in UTC... but at the end of the day you often want to communicate that timepoint to a human user (e.g. an appointment time, the time at which some event happened, etc.), which is when our stupid monkey brains expect the ascii string you are showing us to actually make sense in our specific locale (including all of the warts each of those particular timezones have, including leap second, DST, etc.)
Comment · 1 year ago
> That’s a good target for terrorism.

Exactly! Nobody would be laughing if Al-Qaeda drove a giant Dasani truck into TSA headquarters, would they?

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton