loeg

Are we getting a new Opus 5.1, then?

I totally believe that someone has gotten it this wrong, but personally I've never seen an SMS 2FA on a Passkey authentication. My most common Passkey complaint is that a service doesn't support them yet.

Everything's been discussed to death, including the topics OP chose to focus on. Whatever.

> Every default malloc implementation worked this way about 12 years ago.

Perhaps "default" is doing the heavy lifting here. Glibc malloc was quite bad for a long time, true. But TCmalloc / jemalloc are 21-22 years old, respectively, and jemalloc has been the FreeBSD (released) default malloc implementation for the last 18.

> The solution is and always has been to make larger allocations and use those efficiently.

Having a not-dogshit allocator really doesn't hurt. There's no reason to defend shitty allocator + every application doing manual memory pools on top of it to paper over the bad allocator.

"Using it with a third party high performance allocator allows you to benefit from static linking with very little performance loss" is disputed; 26% is not "very little," even if 144% is worse.

The problem is the "M." Anything beyond a single factor is unnecessarily painful. Make the single factor good (passkeys or FIDO2 or whatever) and the problem is solved without "M."

No mention of memory safe languages? Sure, it does not help existing projects, and sure, of course you can still have logic bugs (or memory bugs using escape hatches like unsafe). But they do help significantly in reducing the number of exploitable bugs.

> Most of musl's performance issues come from their allocator. Using it with a third party high performance allocator allows you to benefit from static linking with very little performance loss.

This is addressed and disputed very early in the article. The very first benchmark presented shows a 26% regression using musl + mimalloc, a high-performance 3rd party allocator.

The 26% number at the top of the article is from using mimalloc (which is a high performance allocator, at least as fast as the glibc allocator) + musl for some task, and the slowdown is coming from (probably) slow musl implementations of memcpy/memset. The musl allocator is even worse.

Smoking's harms are observable in the short term! You don't need long-term longitudinal studies to show smoking is bad. And we have known mechanisms for that harm that relate to combustion. We know mechanically that vapes are not combusting plant material, and we also don't see the same kinds of obvious harms in the short term. It takes willful blindness to the ills of smoking to suggest that vaping might be as harmful or more harmful than smoking.

The article just shows it was too early in May 2026.

Vaping is distinctly less harmful than smoking.

Meta's statement claims $18B. Not sure where Reuters is getting $16.68B from.

https://about.fb.com/news/2026/08/agreement-with-state-attor...

84% is near 90%!

Do you think CGM + alerting is a good solution for this compared with just, like, low-dose GLP-1s?

> Climate change is now a short-term problem.

No, it isn't. There are no short-term solutions.

My understanding is that essentially all high-level endurance athletes are just maximizing their simple sugar intake at all times (during exercise) and knowing blood glucose/ketone levels wouldn't change anything about that intake, nor would it have any impact on crashes.

> Automated glucose control will be the next frontier in so many areas of healthcare.

What areas do you have in mind outside of diabetes management?

> to improve performance

How? Like, what is the mechanism where this is useful?

Methane breaks down over a relatively short timescale (10 years).

> What is the main use case for X these days?

The same as it ever was. Short-form gossip. News, discussion, jokes.

This is one of those situations where describing your algorithmic feed tells us more about you than about Twitter.

I'm not sure I'd agree that the quality is significantly higher than the median literate native speaker. Who do you have in mind? I interact some online with people I meet through sports, and it'd be difficult for me to imagine the LLM output as superior writing. But perhaps that isn't a representative example of the median.

And CHIP, and SNAP. Whatever. You can certainly say social safety net spending isn't exactly 90%. But it's, you know, most of it.

Social safety net programs* (obviously).

A significant chunk of the "defense" spending is VA. Interest kind of shouldn't count? It's the cost of borrowing for our past spending on, you guessed it, mostly social programs.

Something like 90% of my tax dollars go to social security net programs.

The article is interesting even if you disagree with the headline conclusion -- don't skip it on that basis.

No, it has nothing to do with relative power. It is about economic incentives to encourage creation -- the "business models" you deride in your earlier comment.

> The primary purpose behind copyright law is to foster the creation and dissemination of works for the benefit of the public. By granting authors the exclusive right to authorize certain uses of their works, copyright provides economic incentives to create new works and to make them available in the marketplace.

https://www.uspto.gov/ip-policy/copyright-policy/copyright-b...

What do you think copyright is in service of? It doesn’t exist for its own sake.