It’s because, for the most part, it doesn’t actually have access to the text itself. Before the data gets to the “thinking” part of the network, the words and letters have been stripped out and replaced with vectors. The vectors capture a lot of aspects of the meaning of words, but not much of their actual text structure.

Tourist visits.

Yeah, it’s a total fascist shitshow here right now.

https://www.npr.org/2025/09/02/nx-s1-5525984/venezuela-trump-drugs-caribbean

M-x psychoanalyze-pinhead

Introducing the new Amazon Ring Defender–the only doorbell with a live video feed and live ammunition! Don’t like what you see happening on your doorstep? Now you don’t have to be home to defend your castle!

Yeah, I can’t help reading about the Chicago Police Department (CPD) being instructed not to go along with Tom Homan without thinking of the stories about the Homan Square facility in Chicago where CPD would detain people for days without charging them, without giving them access to their lawyers, etc. If we’re thinking CPD are supposed to be some kind of bulwark against fascism, we’re SO fucked.

Edit: To be clear, as far as I know, the shared name is a total coincidence, but it really makes it hard not to think about the history of the department.

I also use way too many em dashes (usually as double hyphens), but I also overuse parentheses and commas and just overly long sentence structures. I would like to think that my style remains pretty distinct from LLM output style at the moment.

The thing that really worries me is that as they stop using weird identifiable quirks like em dashes and emoji, it could be that the identifiable trait that remains is eerily consistent grammar. It used to be that people unconsciously treated extremely grammatical text as authoritative, regardless of its actual merit; as such, teachers spent literal decades drilling into me the habit of avoiding grammatical errors. Now that could end up instead just making folks think I’m a robot, and thus to be ignored.

I guess the actual robots will probably talk less about their neurotic concerns, though, so I’ve got that going for me, which is nice.

Why would you think it’s only foreign agents who would use ICE as cover to disappear politicians they don’t like?

Yeah, I definitely read that as an effort to preempt the folks who were going to yell about how clearly this means the Flipper Zero should be illegal. Hacking has been so poorly represented in TV and films that there are a distressing number of people who don’t realize the term can even have a positive connotation.

Obligatory XKCD: https://m.xkcd.com/1428/

I’m not getting the same error anymore, but the site still seems fucked up. At one point it loaded the desktop version of the old reddit format for no apparent reason on Firefox mobile.

The speculation I saw was that it’s just Firefox in general. Folks were saying turning off ad blocking didn’t help.

For those like me who had to look it up, Bazzite.

See, that makes it sound to me like you could probably come up with a setup that would do what you want, but that doing so would probably mean making it worse at some of the other things you currently use it for.

Which is where using an external drive for a third installation might be easier. Or at least easier to dispose of if you get sick of the project. But I am perhaps unusually lazy in that regard.

Bulgur wheat makes a really good textural element in vegetarian chili.

I think there’s huge variability, but as a gross overgeneralization AMD gpus run Cyberpunk 2077 a bit faster on Linux than Windows, and nVidia gpus run it a bit slower on Linux than on Windows.

If you’ve got a spare usb hard drive you could always install Linux there for a test drive though. You might be able to find a setup that gets you the extra performance you’re looking for.

I got the number from this page: https://ballotpedia.org/Election_results,_2024:_Analysis_of_voter_turnout_in_the_2024_general_election

which lists it as “eligible voters,” which is in turn based on data from here: https://election.lab.ufl.edu/2024-general-election-turnout/

which discusses the methodology:

The denominator for these VEP turnout rates is constructed by estimating the voting-age population or VAP (everyone age 18 and older residing in the United States). Ineligible populations are subtracted from the VAP consisting of non-citizens and felons (depending on state disenfranchisement laws). Eligible overseas voters are added to the national VEP estimates only as no reliable method exists to apportion these eligible voters to states.

So no, it’s not just registered voters. It’s their best estimate of how many people are old enough and not disqualified.

Remember most eligible voters don’t vote.

Voter turnout in the 2024 election was 63.7%. That’s still a lot of folks who didn’t vote, but it’s not “most.”

That was the joke. The link I was replying to was talking about how Trump always says everything will be happening “in two weeks.”