By May 2007, Norwegian newspapers were going nuts about Facebook, with a network diagram showing the connections between celebrities who had Facebook profiles. jilltxt.net/2025/
By February 2004 a US presidential candidate was on Orkut. But he didn't win. jilltxt.net/orkut-stats/
I wrote a paper about links as currency in 2002. Should have gone way more marxist and dystopian on it all. doi.org/10.1145/5133...
I joined Orkut two days after it launched. I was fairly cynical. I found the Friend-rankning creepy. It was the early days of social media metrics, and led to the commodification of friendship. jilltxt.net/rank-your-fr...
We all joined Orkut in January 2004, Google's first go at social media, and not that differenet from Friendster and LinkedIn except it ranked your friends based on how many friends they had. My real friends rapidly became invisible, I wrote in my blog. jilltxt.net/patchwork/
On January 24 2004, two days Orkut launched, I signed up. Orkut was Google's first social network platform, and I was fairly cynical - not much different from Friendster, LinkedIn and Tribe, I wrote in my blog, but creepier: rank your friends. jilltxt.net/rank-your-fr...
Excel and Word started going mm/dd/yyyy randomly for me a few months ago. My computer is set to do regional date and time settings for Norway and I can’t understand why this happens and it drives me nuts. Never used to be a problem. Maybe SSB has the same issue
At snøvær og det å "få heng" nærast tar bort skilnaden mellom kvinner og menn i langrennsløypa er noko underkommunisert. Om det er det som er årsaken, då. Småsure manne-vibes her
First “criminal immigrants”, then just “immigrants”, then practically “brown people”, and now trans people.
As predicted.
“If you tolerate this;
your children will be next.”
Next week's Critical AI Theory Reading Group will discuss David Gunkel's article "The différance engine: large language models and poststructuralism" link.springer.com/10.1007/s001... - Tuesday 17/3 at noon Norway time. uib.zoom.us/meeting/regi...
Maybe you should apply to do a Computational Writing 2026 retreat at the Banff centre this September with
@docmofo.bsky.social
, Lillian-Yvonne Bertrand and others as your faculty mentors? Application deadline 25 March. www.banffcentre.ca/programs/lit...
Conference here in Bergen 15-16 October 2026: "Critical Theory in the Age of Monsters". Abstracts due 30 April. nordiccriticaltheory.w.uib.no/2026/03/13/2...
🥁🥁🥁 Newly out from us today in Science Advances: “Biased AI Writing Assistants Shift Users’ Attitudes on Societal Issues”.
Large Language Models are providing users with autocomplete writing suggestions on many platforms. Could these suggestions shift users’ own attitudes? (spoiler: YES) (1/7)
Good summary in the coverage by @cornelluniversity.bsky.social:
news.cornell.edu/stories/2026...
Grammarly is no longer a friendly tool to help people become better writers, it is now a full on AI-plagiarism machine. I wrote a blog post showing how Grammarly's Expert review feature (discontinued yesterday) presented fake versions of academics without permission. jilltxt.net/grammarly-ge...
I doubt it - I tried to provoke Grammarly into putting me up as an expert and couldn't do it in the 15 minutes I tried. But I found others, including @katecrawford.bsky.social @timnitgebru.bsky.social, Kate Hayles, @n-w-f.bsky.social @kishonna.bsky.social @johnrgallagher.bsky.social.
And then there's the citation finder, which I think is a big part of the reason so many scholarly papers now have irrelevant citations. Here are two examples of suggested sources that do NOT support the claim made in the text. But it takes work to spot how - easier to just "Insert in-text citation".
One of the tools is "fact check" where it aims to find claims in your text that are contradicted. It often messes this up or produces confusing or just outright incorrect suggestions that it offers to just insert into your text.
There's the expert review of course. When I saw that I thought "can this be legal?" Probably not it turns out, it was removed YESTERDAY and there's a class action lawsuit against Grammarly for this. bsky.app/profile/juli...
I signed up for Grammarly a couple weeks ago to check the AI-generated stuff and was horrified. It'd be hard to be a student and NOT just generate everything. And there are still so many confident errors. A silly one: their AI detector recommends "algorithm-generated" instead of "AI-generated".
Lots of folks asking how to join the class. My lawyers' emails and contact-form here. As well as the complaint if you want to take a look.
prf-law.com/current-case...
I saw Kate Crawford, Kishonna Gray and a bunch of other researchers I knew as “experts” in Grammarly. Sorry didn’t screenshot it all but here’s a screen recording me and Kishonna made last week. jilltxt.net/temp/grammar...
New Australia Institute research shows that state and federal governments will provide $16.3 billion in subsidies in 2025/26 to some of the biggest, most profitable companies in Australia, an increase of 9.4 per cent on last financial year.
Read more on The Point: https://theaus.in/4bmdAaP
I took the train to Oslo today, so had time to write up a blog post about yesterday’s AI theory discussion, which was about @ryanheuser.com’s paper on LLM-generated poetry, Jameson, the gimmick, idealisation, rhyme and metre. jilltxt.net/do-llms-norm...
En av de første gangene jeg ble intervjuet tok de et sånt bilde av meg. Året var 2000 eller 2001, og avisen var The Age i Melbourne. Heldigvis ikke på nett. Bildene tok like stor plass som teksten. Ung kvinnelig akademiker for the win.
omg
I've heard librarians & archivists discuss reasons to NOT make every archive/dataset open to becoming training data for AI - can anyone point me to best practices, good arguments etc about this? We're considering ethics around publishing a dataset that someone might use for training. #openscience
In fairness, people were really confused about fictional characters in early novels! It does seem like there’s a parallel there.
