Quekid5

I'm not sure I get your point. Can you elaborate?

We have a marvelous thing in developed countries where the IRS (equivalent) already knows how much your tax should be paying[0] and you basically have to just look at a few numbers and click "That seems right". They even have a form where you can change things that aren't correct! Marvelous, I know... we've had that for 15+ years in Denmark.

EDIT: To add, if you get it slightly wrong... you're fine, it'll be adjusted for in next year's taxes.

[0] ...because it has to, in order to investigate evasion.

> Playing stockfish is like playing tennis against the wall (for untitled players at least).

It's the same for Magnus Carlsen. Even with Queen odds, Stockfish is literally unbeatable for the best players in the world. It's just too strong at evaluating all kinds of random tangent moves (and ensuing positional advantage) which no human player can possibly pay attention due to the time required.

Stockfish vs any human is like Carlsen vs other players by about 3-5 orders of magnitude[0]. It's that stark.

[0] A wild pun appears.

EDIT: To avoid having to respond to each responder, fair comments about Queen odds. Maybe I was thinking Rook odds? Also, I kinda lumped Stockfish in with all the other engines, but I realize there are other engines with different properties ofc.

Chess is a fun game. That's why it's been around for 1000+ years.

There was a renaissance during Covid and due to 'The Queen's Gambit' where it gained much more mainstream popularity, but... Chess AI was already far far (like 1000+ Elo) ahead of human players at that point.

The thing is... chess is humans playing (communicating) with humans and that's what keeps it interesting. Check out the view counts of chess AI tourneys vs. human tourneys.

I mean the OpenAI/HuggingFace thing was incredibly similar, except maybe more reckless than specifically intentional, I suppose. Didn't stop them milking the doomer angle for PR.

I have to I say (if it's sarcasm) that I appreciate the commitment to the bit. Unfortunately, it's getting harder and hard to tell these days...

If the whole British Post Office Scandal (Fujistu, btw) in the UK wasn't enough, nothing will be. People literally killed themselves over the accusations/convictions that were 100% false. One of the major people responsible was an Anglican Priest(!) of all people who was so sure that the Computer Is Right... [0]

We're headed for even rougher times if LLMs are going to be taken "at their word".

[0] Now, watching her testimony, it seems she was willfully complicit and it's a disgrace that she isn't in prison, but that's tangential. The "system" was responsible, but without personal accountability, nothing will change. So it goes.

That's a fair point. Ultimately, incentives are all screwy at the moment.

Gotta tie some actual prestige/shame to not following up.

Absolutely. Very happy 10+ years subscribed here and I'll keep subscribing.

(As someone pointed out in the comments, the title of this specific post could have been less heart-attack-inducing, but still... :) )

Yup, they are often just Accountability Sinks.

Not quite what you're asking, but pre-registration does alleviate e.g. p-hacking and such without the need for multiple teams. It doesn't do much for outright fraud, but still...

Exponential is also overkill (even with jitter as others have mentioned).

I seem to remember there was a "you failed 5 PIN entries in a row, please wait 500000 seconds before you retry" on Apple phones. So, you probably also want a sensible max... which makes exponential a bit pointless. Just do a basic fixed delay + (large, e.g. 0.5 x the delay) jitter and you'll be fine for most things. You can add a bit of cumulative delay if it's really costly to do retries.

LinkedIn was overtaken by LLM nonsense at about the release of ChatGPT 3... I mean it was incredibly stupid and transparent engagement slop before that, but still.

I remember talking to a business type who was so incredibly proud that GPT could turn his incredibly vapid thoughts into bland pointlessness on LinkedIn... not realizing that everyone in his feed was doing the same thing. We never did business with him for ... reasons, obviously :)

It might surprise you to learn that that would also be considered a dick move.

Yep, once people discover the multitude of JVM flags the cargo culting begins...

It really applies to anything with a huge number of options, I remember similar things from back when I used Gentoo and the wild suggestions I'd get for what exact CC/C++ compiler options to use...

Lack of empathy tends to trickle down, alas.

It's definitely less likely with that technique (in practice!), but still very possible, via e.g. ODR violation or even the classic SIOF.

So, ideally you'd want each separate test case to be compiled separately, but even then you wouldn't be safe! ... because any UB in that test (or the code it's testing!) could lead to a random pass.

UB is good in some ways, but other ways it's really really bad.

EDIT: I will say: If you have a UBSAN turned on for testing, etc. you're reasonably safe... but not fully. There's a lot of stuff they don't catch because it's essentially impossible.

Because this is garbage PR. That's it.

Every property-based testing system (invented ca. 1980) will explore boundary values. The semantics (or lack thereof) of C and C++ can make this difficult to actually test for because the compiler is allowed to say "test passed" to any input leading to UB.

Ask for more regulation.

*Brits

EDIT: Don't get me wrong, my European nation did a lot of bad shit, but don't put the idea of an "expat" on us.

AFAIK, the only correlation with LoC that's got solid evidence is this: the number of bugs correlates with LoC.

Must have 120 fps for answers arriving in [buffering] 30 seconds.

Indeed... why is Anthropic even employing people at all if this AI magic story is true?

You misunderstood their example, I think.

If doesn't matter what 'p' is in their example. The point is: if 'f' is undefined behavior (rather than just impl-defined), then the optimizer concludes that the "if p { f() }" can never happen... which means that we're allowed to assume that 'if p { ... } else { ... }' (in the first part of the example) will always take the else branch. The compiler will optimize accordingly and just always call g() unconditionally.

It was released at "end of life" for Larian patches for BG3, so I can imagine they didn't want to support any old Linux distro. I believe it is technically possible to run it on distros that are "similar enough"[0], but it's pretty pointless if you want to use any PC-only mods, etc.

[0] I still use Arch, btw. :)

Yeah, I've been playing BG3 on Linux[0] for about 2 years at this point (using a Lutris "recipe" or whatever they call it). Ironically, the biggest issues have been with some of the modding tools needing specific versions of DotNet and whatnot. The game itself runs flawlessly.

[0] Arch Linux, btw, because that must be mentioned.

The overhead isn't zero, but with SSDs (and filesystem caches in the gigabytes these days) it's damn near insignificant in pure terms of opening files and such.

Generated tests... I mean... listen to yourself.

I can generate a lot of tests amounting to assert(true). Yeah, LLM generated tests aren't quite that simplistic, but are you checking that all the tests actually make sense and test anything useful? If no, those tests are useless. If yes, I don't actually believe you.

It's the typical 10 line diff getting scrutinized to death, 1000 line diff: Instant LGTM.

Pay attention to YOUR OWN incentives.

> However, the best engineers I know are usually among the quickest to open an editor or debugger and use it fluently to try something out.

That's not my experience... mostly it's about first interrogating the actual problem with the customer and conditions under which it occurs. Maybe we even have appropriate logging in our production application? We usually do, because you know, we usually need to debug things that have already happened.

(If it's new/unreleased code, sure fine, let's find a debugger.)

Wikipedia says it was created in 1974, so something's definitely off with the time line here.