These respondents were self-selected Guardian readers. Doubt many voted for Trump.

Kinda reminds me of this: I built the most expensive CPU ever! (Every instruction is a prompt)

I’ve always seen them curved like this:

But I suppose they could be straight; here’s one that looks like it just uses a chisel (doubt it would work as well though):

Ain’t broke and I can’t be bothered to update. Not accessible publicly either. It also runs some software with very specific and brittle dependencies and I don’t care to risk breaking it. If distro is EOL (probably is) then it’d be a pretty time consuming getting everything set up again.

When something doesn’t work. I.e. when an app update causes incompatibility with a service. I think I have one server that’s a few years without an update (distro version may actually be EOL for all I know).

You can plug a PC into a TV and even use Xbox and PS controllers if you like. Can have the PC auto start Steam in Big Screen mode so you’d seldom need to use a mouse+KB (trackballs or keyboards with trackpads are best in this scenario). Nothing wrong with console gaming though (well, besides supporting all the non-free software, locked-down systems, and shit companies; but most games you likely play are these as well).

As an American, since I was a teenager, I always thought Canada or the Scandinavian countries would be better places to live (mostly due to cannabis laws and healthcare). I grew up in rural US, so was aware of high levels of bigotry and poverty, but yeah, it seems to be getting worse. Rights have been getting eroded since W (except LGBT rights, until MAGA). The economy, for the working class, has been in decline since I’ve been alive. We seem to be going through a corpo-fascist self-coup now, so it’s definitely worse. In terms of foreign meddling, I don’t think much has changed since I’ve been alive until this Trump term. Disregarding LBGT acceptance, I think the late 90s were peak. Regarding WASP men, I think the 50s may have been the peak. The country was built on genocide and slavery in many uniquely brutal ways, so don’t think it was really “great” until the mid 1900s, but not even sure it was comparatively great to other peer countries back then, in anything other than having a large economy and plentiful resources to be exploited.

Yeah, that could also provide an incentive for companies to produce stuff in ways that reduce carcinogens, yet still have some amount. I think traditional bacon that doesn’t use synthetic curing salts contain less nitrates, for example.

Venezuela isn’t even a major source of cocaine or even a major transit route for it.

I think abstract support of genocide should probably be legal, I don’t think organizations, institutions, or people should be forced to tolerate it though. Anti-fascists being anti-free-speech in regards to pro-fascist speech seems ok and natural?

Excessive use of em-dashes, emojis, and other characters that aren’t on standard keyboards. I think these companies purposely have the models generate this stuff so it is easily detectable (so they avoid training on their own slop).

If I understand the results tables on repo correctly, their largest model achieves ~68% re-executability rate on code compiled with the q0 optimization flag. I’m unsure if that just tests if the decompiled code can be recompiled and executed, or if the programs need to produce the same result on some test cases. If the model is used to refine Ghidra outputs (I’m guessing this is some well-known decompilation framework) it can be used to achieve ~80% re-executability rate, which is better than Ghidra’s baseline of ~34%.

They often operate on the “just-world fallacy” too. I.e. if people are poor, starving, arrested, deported, raped, it’s because they deserve it.

https://www.wired.com/story/ice-social-media-surveillance-24-7-contract/

Seems to be more of a tutorial for the optimizer. Would’ve been more interesting if they delved into the problem and related problems. I think this problem could be reduced into other well known problems with optimized algorithms (seems related to the knapsack problem).

Working with/on things I found interesting helped a lot. I.e. lots of small projects/scripts, using different frameworks/libraries/languages that looked interesting. Experimenting and exploring different ways things could be done. Programming is one of those “10,000 hours” things; you need to be interested in what you’re doing to do something like that for so long. Computer Science coursework helped a lot too, especially the courses heavy on algorithms, data structures, big-o, proofs, etc.

In my CS coursework, we were exposed to many different languages and programming paradigms at the very beginning. It’s fine to experiment and start learning multiple languages at once (preferably, all being quite different, such as a pure functional language, procedural language, object-oriented, declarative logic, etc).

Outside of the “stars” and directors, people working in the film/tv industry already have a fairly low income (median salary for professional actors is ~$47000/yr).

I started using it as an alternative to Octave/Matlab and Perl. Python is better at general programming than Octave/Matlab, and better syntax than Perl (IMO) while being almost as easy to do the same stuff I was using Perl for. It’s very good for quickly writing small scripts. Issues can arise on large projects/teams because of stuff like type safety, and it also has issues with performance.

I think compose files are usually pinned to a version, or use a .env file that needs to be changed to update to a new version.

I personally don’t update very often; usually not until I’m forced to for some reason. I find that just checking the documentation for any upgrade/migration guides, and doing it manually is sufficient. I don’t expose this kind of stuff publicly; if I did, I’d probably update regularly.

Makes it hard to use h1bs the way they were supposedly meant to be used; to make it easier to hire talent that can’t be found domestically. It think this will accelerate offshoring even more. I would have preferred the program was just reformed to make gaming the “prevailing wage” requirement harder, and to give h1b workers more freedom so they’re not as easily exploitable.