Yup, your LLMdar is better than mine, as the other reply to me links to the author admitting it on HN. I'd like to be better at detecting it, but if the price is having to read loads of mediocre prose then the price is too damn high.

Is it? I know there's ai in the domain, but I don't get that vibe from the style, and I just checked out the claimed authors and am confident they're real people who've been writing since before LLMs could do it well.

Good news for red squirrels!

That gun... is this a mod/censor-friendly setting, or is that a genuine weapon from the game?

A couple more things - myself and the mods can’t influence a live game in any way.

I joked several times about that, but I absolutely didn't believe it - it was just funny that every time you applied an update you pulled a key or two.

It would’ve been neat to be able to see who used the reactions in chat. (Like, hovering over them I’d expected to see the names of people who selected it rather than just the count.)

I was thinking that. However, mobile users have to click to get the hover effect, but that already toggles you using that reaction. I think hover + a separate button with a modal would work.

I like it, and its scaling difficulty is smart.

HexHunt #13 · 610/1400 · top 22%

🇯🇵 🇨🇳 🇸🇭 🇰🇪 🇵🇾 🇪🇬 🇧🇷
3 first-guess · 13/21 guesses used

Wow, it's like if the Phoronix forums didn't have its sane element. Is this representative of Artix users generally?

Please reread the first sentence of what you just replied to.

Cloudflare gives its 522 error, but I can load via IPv6 from the same city. I've also found the issue with Cf in front of servers I run on a different continent, so I've been able to do some good testing and have spent plenty of time looking into it. I haven't found any other Cloudflare locations with the problem, but there's plenty of them.

Personally, my local Cloudflare seems to have some problem with IPv6 and some sites won't load. On my desktop I've set Firefox to use IPv4 for such sites, but on my phone I need to use mobile data (which doesn't support IPv6). I don't think it's a common problem, but you did ask.

Gotta love Ralph Steadman. I've got a copy of his Extinct Boids book.

I don't think having a textual article auto-scrolled and voiced adds anything at all, and people who do want that can do it themselves, and have it set to their choice of accent, speed, etc.

Video links tend to do worse than text links on Lemmy, so you'd probably have been better off all round just posting your original blog post.

This is just a bad text-to-speech reading of a Medium blog post.

Edit: I see the author has satire in the list of tags, so perhaps if I'd stuck with it longer I'd find it's a joke...?

I made a rubber duck for all your debugging needs:

That elongated, streamlined shell made me assume this was a joke or hoax at first.

I hadn't heard about the carrot disclosure stuff, but I hope this doesn't become a common thing. Just saying "I can hack you" with some purported output from a script is pretty useless feedback.

Geoffrey Challen, a computer scientist at the University of Illinois at Urbana-Champaign, plans to offer a new course this fall in which he will teach students to develop software “without writing, reading, debugging, or viewing a single line of code,” he told me.

Is that meant to say reviewing? Either way, I can't see how this would lead to good results, even with a comprehensive test suite. Security? Scalability? Maintainability?

Have they taken out the AI generated papers? We know that training LLMs on LLM-generated text leads to an absolute collapse in quality, and we also know that AI has been showing up in papers so if they haven't, then this will be quite unreliable.