On my system that consistently gets results around 10s and 5s so the difference is sort of interesting. Mine’s a Ryzen 3600, maybe newer CPU features are of substantial benefit to xz.

The .tar.xz format decompresses more than twice as fast as .tar.bz2, allowing you to get up and running in no time

$ time tar xjf firefox-134.0b3.tar.bz2 

real    0m9.045s
user    0m8.839s
sys     0m0.450s

$ time tar xJf firefox-135.0a1.en-US.linux-x86_64.tar.xz                                                

real    0m4.903s
user    0m4.677s
sys     0m0.510s

Nice! Presumably it’d be twice as fast if disk was infinitely fast or something. Unfortunately by testing this I’ve already used up a hundred times more time than I’ll ever save as a result of it.

There are currently zero bluesky servers that aren’t fully dependent on bsky.app.

the vast majority—up to 82%—of the links to false information in the dataset originated from conservative news domains.

I’m not sure what approach would work. As I understand it, it’s designed around the idea that all messages get routed through a monolithic “relay” which needs to see every single event from every user in order for any of them to get routed between the PDS nodes where user data gets stored.

Probably best to just add ActivityPub on top of it, if they really wanted to federate with anyone.

Just a hunch, but I think no decentralised network is going to run ATProto unless someone other than Bluesky forks it and makes the protocol changes that would allow that to be practical. I guess it’s possible, in theory?

When I looked at bsky it seemed to be about 60% pointing out stupid shit that Trumpists have recently said and done, and 30% self-congratulatory talk about how popular bluesky is this week.

the study suggests that individuals who deviate from their party norms are quickly treated as if they are a political enemy.

Is that “party” as in political party? Because I don’t know about the rest of the world, but in Canada it seems like the main polarization is between the Conservatives who have their shiny new Conservative party on one side, versus everybody else who doesn’t really have an official party they identify with all that much on the other side. It’s not yet like the USA with its seemingly-eternal two-party system. I wonder if it looks like that if you view it through Twitter.

Factiverse’s success rate is around 80%

That seems pretty close to useless even if we assume that the criteria they’ve used to define success are spot on. Funny how they don’t mention it until the second to last paragraph.

And then there’s Lemmy, where you can always count on some helpful stranger who’s completely missed the point chiming in to tell you why you’re wrong.

That’s not how it was as of yesterday when I signed up just to see what all the fuss was about. (I don’t think I’ll be participating.)

Bluesky: You are immediately and automatically welcomed into the warm embrace of an algorithm that entices you into a parasocial relationship with the synthetic community it has created.

Mastodon: If you’re lucky you’ll stumble across a warm welcome for new users explaining how posts are called toots here, likes are called florps, and our version of Grok is called Garfiald.

It’s the culture of an instance that makes the difference, not which software it runs, but there is often a correlation. Misskey tends to get more people who appreciate cute emoji and comfy vibes.

I thought maybe it was just my imagination that it’s been really slow since Wednesday, but you can see it clearly on the charts at the bottom of the page there.

Having apps that do what users want but try to hide it from reviewers really highlights the absurdity of letting Apple decide what software you’re allowed to run.

Cross-country skiing is fun! I think so, anyway. Take it easy, don’t expect to be super good at it right away, et cetera, but maybe try out whichever winter sports appeal to you. What everyone else said, but also don’t forget to go outside and enjoy yourself sometimes.

The Featured Snippet quoted an article from the Mayo Clinic, highlighting the words “Caffeine may cause a short, but dramatic increase in your blood pressure.” But when she looked up “no link between coffee and hypertension”, the Featured Snippet cited a contradictory line from the very same Mayo Clinic article: “Caffeine doesn’t have a long-term effect on blood pressure and is not linked with a higher risk of high blood pressure”.

On the one hand, Google sucks. On the other hand, if people are unable to a) understand how those two snippets are not contradictory, and b) read at least one very short simplified-for-laymen Mayo Clinic article about the topic before thinking they’ve learned anything at all about medicine, it’s hard to see the problem as being primarily due to Google. There is something deeper, and worse, going wrong when people habitually take that kind of extreme shortcut to thinking that they know the right answer about almost anything, and it has little to do with whether any one-sentence snippets they’re given are biased or accurate.

Overclocking

Just think of all the juicy benefits of replacing journalists with machines. They’ll never stumble or cough while presenting the news, they’ll never call in sick, never age, never get mad as hell and decide to not take it any more, never resign from the editorial board in protest no matter what garbage you tell them is the news. Machines are just better suited to the job, it’s inevitable.

GPU has fallen off the bus.

If you’re lucky, it’s an nvidia driver problem. If you’re not, it’s a hardware problem.