James Eagan

232 Followers
184 Following
544 Posts
Researcher, teacher, builder of bits at Télécom Paris, Institut Polytechnique de Paris. Human-computer interaction, information visualisation. Opinions mine. Pardon my 🇫🇷.
Webhttps://james.eagan.fr
Labhttps://diva.telecom-paristech.fr
"The computer science undone: How the social construction of disciplinary boundaries and disciplinary hierarchies shape a field" by @Felienne
https://www.felienne.nl/paper-the-computer-science-undone/
Paper: The Computer Science Undone

How The Social Construction of Disciplinary Boundaries and Disciplinary Hierarchies Shape a Field

Felienne
@pg @amyjko teaching a 16-week class during sabbatical is definitely an interesting challenge when one is used to an 8-week quarter. Also: MWF classes instead of one or two half-days a week.

"HyperDoc - a hypermedia substrate for knowledge workers"

My contribution to yesterday's "Substrates 2026" workshop.

http://hyperdoc.wiki.khinsen.net/view/welcome-visitors/view/substrates-2026-workshop/view/hyperdoc---a-hypermedia-substrate-for-knowledge-workers

hyperdoc a hypermedia substrate for knowledge workers

In early 1985, Henry Nussbacher sent a letter to all Bitnet contacts calling for admins to hunt down and destroy every chat server because a dozen users actively chatting could bring file transfers to a halt—file transfers being the primary purpose of the network.

Jeff Kell realized that the problem wasn't chatting itself, but redundant traffic from independent connections. So he built the first Relay chat system. I got involved in August, setting up the 10th relay server and contributing code.

No, “AI” is not a Stochastic Parrot 🦜 | by Margaret Mitchell | Mar, 2026 | Medium
https://medium.com/@margarmitchell/no-ai-is-not-a-stochastic-parrot-a99e57766bed

Lisez ça, ça remet bien les idées en place.

Mitchell revient sur la métaphore des perroquets stochastiques. Elle réexplique (l'article originel était pourtant assez clair) que son utilisation permet de pointer les pièges anthropomorphiques des LLM.

Elle détaille néanmoins beaucoup plus pour répondre aux critiques récentes.

"LLMs produce fluent language. Fluent language is what humans associate with intelligence. Therefore, the argument goes, LLMs are intelligent. This is an error in logic that the framing was designed to name."

No, “AI” is not a Stochastic Parrot 🦜

I’ve recently come across a new flavor of AI denialism making the rounds.

Medium
Best thing I've read about AI and jobs. TL;DR: "The ATM tried to do the teller’s job better, faster, cheaper; it tried to fit capital into a labor-shaped hole; but the iPhone made the teller’s job irrelevant.”
https://davidoks.blog/p/why-the-atm-didnt-kill-bank-teller
Why ATMs didn’t kill bank teller jobs, but the iPhone did

There's a lot more to replacing labor than just automating tasks

David Oks
Another victory against the vile metric system

New #CHI2026 paper: “How are Vibrotactile Experiences Visually Represented? A Taxonomy of Illustration Characteristics”

We analyzed hundreds of images from research articles to uncover how vibrotactile experiences are visually represented and synthesized their features into a comprehensive taxonomy.

preprint: https://hal.science/hal-05488387/

This work received a Best Paper Award at #CHI2026

#haptics #tactileDesign #visualization #taxonomy

RE: https://mastodon.social/@mdn/116183060176876798

“Works on all major browsers” since 2024. Are you ok with structurally excluding roughly 18% of desktop internet users? (Source: https://caniuse.com/wf-array-group)

I can’t believe I never before thought to use a roast baster when soaking French toast.