Notices by Hugo (hugo@assemblag.es), page 3

threads is now blocking mastodon.social, which—if you have been following the political punch and judy show around this over the last year or so—is objectively funny.

https://www.threads.net/moderated_servers

a lot of this in my feed right now. a great disturbance in the force.

downright hallucinatory to read an entire article like this with not a single mention of covid itself causing immune dysregulation.

https://www.bloomberg.com/news/features/2024-06-14/why-is-everyone-getting-sick-behind-the-global-rise-in-rsv-flu-measles

"ingestion" is a weirdly corporeal, metabolic metaphor to use when scraping data from a network for your new website.

also a pretty clear red flag.

I suppose there's a sort of digital body horror undercurrent to this.

machine learning models consume your data, metabolize it, turn it into parts of their "body" that can't be easily extracted again.

been updating my robots.txt files, thinking it'd be nice to have something like that here: a machine-legible way to clearly mark legitimate and non-legitimate uses of your data.

serious question: if you're an unhappy academic who dreams of quitting your job and living another, better life—what do you do in that other, post-academic dream life?

I ask because academia has a way of narrowing your horizons, making it feel like it's the only thing you can do, the only thing you're good for, your only way to live. always interested to hear what people dream of doing instead.

fresh coffee, birdsong and the soft whirr of a robot lawnmover butting repeatedly into my lawn table.

jesus christ.
https://www.aljazeera.com/news/liveblog/2024/5/27/israels-war-on-gaza-live-news-35-killed-in-rafah-tent-bombings

so by default, anything you do that appears on the screen of a windows user (chat or messages, say, or video calls) will now be recorded, machine transcribed or summarized, and available to anyone who gains access to the folder on that users computer where windows stores these recordings.

@seanbala

yeah, the UK is particularly abysmal on this. a sort of neoliberal laboratory for dehumanizing academics.

fabricated cell lines.

https://www.science.org/content/article/hundreds-cancer-papers-mention-cell-lines-don-t-seem-exist

wonder sometimes what science would look like if we just... dropped peer-reviewed publications as a metric of achievement. came up with better ways of evaluating merit, more humane ways for choosing who to hire or how to award grants.

maybe the peer review system as it exists could actually be redeemed, if the raging torrent of articles slowed to a trickle because peer-reviewed publication wasn't a matter of life or death for everyone anymore.

very logic. so answer. much helpful. so glad my university encourages everyone to use ChatGPT (ChatGPT-4 in this case) as a research tool.

ChatGPT-3.5 meanwhile has a stroke trying to answer this.

I sometimes think about these norwegian meteorologists in 2018, registering the first ever tropical night in tromsø—well past the arctic circle—and celebrating the event with buns and currant juice, posing for a photo in the national news. like complete fucking morons.

https://www.nrk.no/nordland/her-feirer-meteorologene-historisk-varmerekord-1.14145892

not gonna lie, the schismatic beef between jack dorsey and bluesky is making me a lot more favourably inclined towards the latter.

"very, very common people" is an incredible line.

the full interview is here, in all its ridiculousness. https://www.piratewires.com/p/interview-with-jack-dorsey-mike-solana