arthur2e5

> Meanwhile in just about every other area that uses fp computation it's been the defacto standard for decades.

Not that strongly for more parallel things, quite similar to the situation with atomics on cuDNN. cuBLAS for example has a similar issue with multi-stream handling, though this can be overcome with a proper workspace allocation: https://docs.nvidia.com/cuda/cublas/index.html?highlight=Rep....

Still better than cuDNN where some operations just don't have a reproducible version though. The other fields are at least trying. DL doesn't seem to be.

On that note Intel added reproducible BLAS to oneMKL on CPU and GPU last year. https://www.intel.com/content/www/us/en/developer/archive/tr...

Hey! I actually appreciate this kind of HN humor. As in, when you pretend to read the title badly.

The author gives a godbolt link. It takes 5 minutes to add two compilers on raptorlake and see that it gives the same result.

https://godbolt.org/z/oof145zjb

So no, Haswell is not the problem. LLVM just doesn't know about the dependency thing.

Fil-C works because you recompile the whole C userspace. Unsafe Rust doesn't do that... and for many practical purposes you probably want to touch the non-safe-version of the C userspace.

Still, it's all LLVM, so perhaps unsafe Rust for Fil-space can be a thing, a useful one for catching (what would be) UBs even [Fil-C defines everything, so no UBs, but I'm assuming you want to eventually run it outside of Fil-space].

Now I actually wonder if Fil-C has an escape hatch somewhere for syscalls that it does not understand etc. Well it doesn't do inline assembly, so I shouldn't expect much... I wonder how far one needs to extend the asm clobber syntax for it to remotely come close to working.

HTTPS is there, so you go down to that level only if you want to distrust any element of the public key infrastructure. Which, to be fair, there are plenty of reasons if you are paranoid -- they do tell you who's doing what in a shady way as they revoke, so there's a huge list of transgressions.

AM404 really is the metabolite that keeps giving. The central nervous system system effects were fun too, acting on cannabinoid receptors and/or TRPV1 channels -- so either stony or spicy.

From my limited experience of metamizole it feels a bit stronger than paracetamol/acetaminophen. Neat little drug if your genetics can take it.

Tangential: China technically banned metamizole due to the agranulocytosis scare, but somehow small clinics always have fresh stocks of this stuff. And their stocks don't look like my metamizole for horses! It's pressed out of the usual magnesium stearate instead of whatever rock-hard thing they use for animal drugs in China.

It does sound similar. Throw in a few helping lines about understanding why you do the "thinking" and it might even become dialectical.

Not a professional, but still use it like that. They also have a new smartblast thing, which works much faster (really, really like Google!) but only on highly similar proteins.

on one hand i can see how it's going to break stuff (and I see they've changed it! []). on the other hand... mind talking a bit more about how it did break things?

[]: https://github.com/boostorg/container_hash/commit/40ec854466...

There's some work in looking at alternative ways of administering the BCG vaccine (iv injection, inhalation). The IV numbers are pretty promising -- if the monkey numbers hold. I hope that someone, somewhere pushes it into human trials soon.

There's also the nebulous but very nice-sounding "non-specific effect" with it. Who knows.

Really bad nitpick, but: I cringe at FOO'S-type plurals. Just break out of the uppercase and say VHDXs if you find VHDXS too confusing.

https://www.hamilton.edu/academics/centers/writing/seven-sin...

We do know that for the lucky ones on which SSRIs do work, it seems to reverse some of the brain structure changes associated with depression -- some abnormally small parts tend to grow thicker. The reversal can be seen at one week after the start of medication use, before one can even tell whether an antidepressant has started working (a commonly-cited duration would be 2 weeks).

[1]: https://www.nature.com/articles/s44220-023-00187-w

resistance training is indeed the recommendation for non-head bones, if you’re looking for a confirmation:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6279907/

Lenore’s approach uses rejection sampling, which translates to needing an uncertain amount of input — a uint128 may or may not be enough.

But there is one related algorithm that doesn’t: https://github.com/apple/swift/pull/39143. This method only requires (output word size + 64) bits, which is really convenient for us since we probably don’t have 2^64 buckets. Same ballpark as (hash >> 64) * n, just with bias correction.

With this method the output value will be very not-scrambled relative to the high bits of the hash. Now whether that’s a problem is… someone else’s question. One can always do some shifts and XORs.

No known consequence, assuming...

(1) changing the RSA decrypt function in OpenSSH is all the code hidden in crc64 does: that's the only known behavior, but we don't know what the changed function does besides letting some authentication through, nor do we know if there are other things it does

(2) there's no malicious machine in your LAN exploiting the RSA decrypt to log onto your sshd: nobody has seen one yet, but it doesn't mean there's no such thing.

If you are not using a distro that does dpkg or rpm, or if your machine is not x86-64, you're free from the "code hidden in crc64", the one that targets sshd, CVE-2024-3094. Are there unknown backdoors? Who knows. Do we count the landlock sabotage as a backdoor?

It's hard to deal with unknowns. Assume the worst, maybe, but what even is the worst?

I don’t know how to break it to you, but cell membranes are not blood vessels. Things go across… differently.

The delayed access "embargo" scheme is not that bad… if you are not chasing the latest stuff in the field. It works well enough for casual Wikipedia editing and junior researchers.

I don’t deal with the ASA much, but I do use the PNAS a lot — their HTML paper layout isn’t bad and they do maintain a nice archive. The National Academy of Science deserves some money for good web work, I just don’t know whether it’s just for them to take it from the embargo.

Also, how much marginal utility does "the public" gain when you reduce or remove an embargo?

Macfuse itself has been closed source for a long time. Whether their EULA is compatible with the LGPL though, I don’t know. (Also, the old question: are APIs copyrightable?)

India is the world’s bigger producer of turmeric and many other spices after all. Won’t be surprising that these batches originated in India and were made using adulterated rhizomes.

A typical "not-so-poor" country with large scale nutrition problems would be China. A look into the micros shows that ALMOST EVERYONE is short on calcium, riboflavin, a couple of other B vits, and Vitamin A. The current deficiency data by DALY accordingly shows large disease burdens on iron deficiency (somewhat linked to riboflavin), good old protein-energy malnutrition (child stunting and the like), and Vit A deficiency.

A lot of the issues with micros can probably be solved through food fortification, except the public is not known to appreciate the price increase (however small it will be) and the longer ingredients list. You might also suggest switching from white rice to brown or parboiled rice, but economics of scale and (for brown) the need for airtight storage mean that these things are much, much more expensive than a 2-year-old sack of white rice.

Even then you haven’t solved the big protein-energy issue. Sure, push legumes more, but these things still don’t have the price-calorie ratio that rice has, and you might compromise the energy part as you do it.

(For non-kids there’s still a big protein issue. Laborers just want to eat enough fuel to keep working, and good old white rice with Lao Gan Ma wins every time. Tofu is fast to shove down too, but it’s mostly water.)

Micros: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9696081/

DALYs: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9570758/

On the cold, industrial side where people just do "close enough", Penicillium roqueforti is also used to make concentrated blue cheese flavor. Food scientists know that blue cheese flavor mainly comes from ketones produced by the mold, which in turn derives from milk fat. So they start with cream, mix the spores in, then aerate and stir to let the mold eat as much fat as possible. 48 hours later, you get what salad dressings use for "Blue cheese" flavor.

The article doesn't specifically require mold; it only argues that mold is acceptable if the cheese is hard and salty. It's more about how cheese that's been forgotten in the fridge actually continues to age and develop flavors -- a poor man's cheese-aging cave.

The big fun tidbit is the 40-year-old cheddar from Wisconsin, forgotten in a walk-in cooler.

Ditching Rosetta 2 would mean a lot of application support going down the drain (including Crossover/Wine), but such things really haven’t stopped Apple before. On the other hand the Linux VM support at least shows that Apple has interest in building on top of Rosetta 2.

Windows arm64 has a very interesting approach to the same problem: the ARM64EC ABI provides 1-to-1 mapping between x64 and ARM registers, so that an x64 program can link to a ARM64EC library. It blocks you out from additional registers sadly.

These polyfills aren't just large, they're also slow because they never use the native implementation. The good news is that you can replace them all with "overrides" in package.json. That's what nolyfill (https://github.com/SukkaW/nolyfill) does. Oh, and of course the README mentions ljharb.

There is an "engine"!entry in package.json, and a package manager should be smart enough to not upgrade past what it can. But people tend to not use that sort of dependency flexibility and use lock files…