{ "jason": "clark" }

A broader vision of open research is needed to include the arts, humanities and social sciences - LSE Impact Discussing the findings of a new report on open practices in the arts, humanities and social sciences, Jenni Adams, Miranda Barnes, Samuel Moore and Stephen Pinfield argue to support open research in…

A broader vision of open research is needed to include the arts, humanities and social sciences.

@jenniad.bsky.social @mirandab-oa.bsky.social @samuelmoore.org & @stephenpinfield.bsky.social for @lseimpactblog.bsky.social 👇

This is a dramatic update to the NIH Data Management and Sharing Plan template, effective for applications submitted for due dates on or after May 25, 2026.

grants.nih.gov/grants/guide...

Props to @scott.hanselman.com for naming the elephant in the room in big tech.

AI coding tools boost senior engineers but create challenges for juniors who lack context to judge outputs. Cutting junior hiring for near-term gains weakens the future pipeline. We’ll need apprenticeship-style paths.

Image of a man (Agent Cooper) driving and talking into a voice recorder. Closed caption reads: “Diane, 11:30 AM, February 24th. Entering the town of Twin Peaks.”

Happy Twin Peaks day to all who celebrate!

What do LLMs see?

I wrote a lil' tool that extracts the attention matrices out of open models and creates this typing visual, with each token's opacity changing according to its average attention score as the prompt progresses. Dimmer words are considered less important to the model.

GitHub - DrCatHicks/learning-opportunities: A Claude Code skill for deliberate skill development during AI-assisted coding A Claude Code skill for deliberate skill development during AI-assisted coding - DrCatHicks/learning-opportunities

Key to efficient learning is realizing how we ACTUALLY learn, not just what FEELS like learning. I wrote a Claude Skill for some friends to help them think about this and they've liked it -- see Principles for some directions you could explore

github.com/DrCatHicks/l...

RDA-US is launching a funded program for US-based professionals working in/with research infrastructure. Great professional development and networking opportunity...and great way for newcomers to engage with the Research Data Alliance (RDA)! rda-us.org/announcing-t...

Wikipedia at 25: What the web can be - Anil Dash A blog about making culture. Since 1999.

"Whenever I worry about where the Internet is headed, I remember that this example of the collective generosity and goodness of people still exists." anildash.com/2026/01/15/w...

Where CC Stands on Pay-to-Crawl - Creative Commons Creative Commons proposes principles for responsible use of pay-to-crawl that prioritizes openness, reciprocity, and the commons.

“Pay-to-crawl refers to emerging technical systems used by websites to automate compensation for when their digital content—such as text, images, and structured data—is accessed by machines.” @creativecommons.bsky.social

Intertapes — Main Obscure tape finds and their stories

Found the coolest website that takes random found cassette tapes people submit and digitizes them. I’m listening to an NYC hip hop station from 1994: intertapes.net

Full moon over a frozen lake. A small fish house on the lake.

First full moon of the year.

Is ‘open science’ delivering benefits? Major study finds proof is sparse It’s hard to measure social and economic impacts of making papers and data free, researchers say

"Does the open science movement—the push to make research outputs such as articles, data, and software free to read and reuse—produce the benefits its supporters claim, such as accelerating discovery and promoting science literacy? The answer is a qualified yes."

The year of “reasoning” The year of agents The year of coding agents and Claude Code The year of LLMs on the command-line The year of YOLO and the Normalization of Deviance The year of $200/month subscriptions The year of top-ranked Chinese open weight models The year of long tasks The year of prompt-driven image editing The year models won gold in academic competitions The year that Llama lost its way The year that OpenAI lost their lead The year of Gemini The year of pelicans riding bicycles The year I built 110 tools The year of the snitch! The year of vibe coding The (only?) year of MCP The year of alarmingly AI-enabled browsers The year of the lethal trifecta The year of programming on my phone The year of conformance suites The year local models got good, but cloud models got even better The year of slop The year that data centers got extremely unpopular My own words of the year That’s a wrap for 2025

Here's my enormous round-up of everything we learned about LLMs in 2025 - the third in my annual series of reviews of the past twelve months
simonwillison.net/2025/Dec/31/...
This year it's divided into 26 sections! This is the table of contents:

Ok a translation is up.
No matter what your take on these rules, it's hard not to admire China trying to get ahead of these pressing issues: data protection, dependency, labeling and reminders that you are talking to a machine, etc.
Plenty to unpack here.
www.chinalawtranslate.com/en/chatbot-m...

In a new blog post, I contrast two flavors of empiricism: the one practiced in the social sciences and the one practiced in ML/CS.

I argue that we need both, given that CS is increasingly about "claims," and not just constructing artifacts.

doomscrollingbabel.manoel.xyz/p/the-empiri...

IIIF Illustration Detector - a Hugging Face Space by small-models-for-glam Find illustrated pages in digitized historical books

Paste any IIIF manifest → model classifies every page locally → see where illustrations appear.

Part of small-models-for-glam: small, efficient models for cultural heritage work.

Not everything needs GPT-4!

Try it: huggingface.co/spaces/small-models-for-glam/iiif-illustration-detector

‘Sam’s the biggest Cooke in town,’ New York City, 1964

If you are interested in the crawl to referral stats I mentioned, here's the Cloudflare blog post - blog.cloudflare.com/ai-search-cr...

Data is from the Cloudflare AI insights tool - radar.cloudflare.com/ai-insights#...

small-language-models-trust-clark-ff2025-shared Reasoning with Small Language Models (SLM) to Create Trustworthy GenAI Jason A. Clark @jasonclark Montana State University Fantastic Futures 2025

Slides and code from my #ff2025 talk, "Reasoning with Small Language Models (SLM) for Trustworthy Generative AI (GenAI)."

Slides:
docs.google.com/presentation...

Code:
github.com/jasonclark/a...

Meet the Librarian Resurrecting the Lost Women of Skateboarding Female skateboarders have always been there. Someone just needed to prove it

The kind folks at The Walrus published this feature over the weekend! Thank you! #skaterlibrarian thewalrusca.substack.com/p/meet-the-l...

Introducing the 'Software Paper': New ways to publish on research software in Computational Humanities Research When Lauren Tilton first approached me about joining the Computational Humanities Research (CHR) journal’s Editorial Board as an Associate Editor, the thing that made the invitation so compelling and ...

“The Software Paper fills a gap for the computational and digital humanities communities...” Thank you to research software engineer extraordinaire @suttonkoeser.bsky.social for leading this initiative for Computational Humanities Research journal. Please share!

Thinking of Anne-Wil Harzing 12 guidelines for good academic referencing & how Generative AI engines even RAG/Deep Research often breaks many/most of them (1) harzing.com/blog/2016/04...

A systematic examination of generative artificial intelligence (GenAI) use guidelines in applied linguistics journals The unannounced appearance of GenAI in 2022 and the speed of its adoption by researchers have left many questions unanswered about its accepted ethica…

For academic writers, editors, and publishers looking for a systematic overview of emerging GenAI policies in an academic field, please see Yin and Chapelle's (2025) excellent paper which provides this for the field of applied linguistics: www.sciencedirect.com/science/arti...

A nine-panel cartoon grid titled "CONFLICT IN LITERATURE" depicts different types of conflict, categorized into "CLASSICAL," "MODERN," and "POSTMODERN" literary periods. Under "CLASSICAL," panels show "MAN vs. NATURE" (a person with a pitchfork against a pack of wolves), "MAN vs. MAN" (two people in a sword fight), and "MAN vs. GOD" (a person pointing a stick at a giant hand from the sky). The "MODERN" column shows "MAN vs. SOCIETY" (a person standing alone against a crowd with protest signs), "MAN vs. SELF" (a person facing a mirror image of themselves), and "MAN vs. NO GOD" (a person pointing a stick at an empty sky). The "POSTMODERN" column depicts "MAN vs. TECHNOLOGY" (a person with a baseball bat facing an array of computer screens), "MAN vs. REALITY" (a person standing in a surreal landscape of melting colors and shapes), and "MAN vs. AUTHOR" (a person standing on an open book, looking up at a giant hand holding a pen). The artist's signature, Grant Snider, is at the bottom right.

Interesting. The evolution of conflict in literature / society

(by Instagram user: @grantdraws)
www.instagram.com/p/DBOlpAuRj9c/

The Best Album of 1989 Round 5 Match #124: De La Soul vs. Fugazi This is the Best Album of 1989 match for Thursday, November 20th, 9am CST. It will be open for 24 hours, and the winner will move onto Round 6 (Semifinals) of the bracket. If you haven’t heard either ...

The Best Album of 1989 Round 5 Match #124
#3 De La Soul, 3 FEET HIGH AND RISING
vs.
#11 Fugazi, 13 SONGS

forms.gle/R6JGyPyp7DQ8...

Responsible AI in Libraries and Archives team has released our interactive Viewfinder toolkit.

Toolkit (interactive website): www.lib.montana.edu/responsible-...

Project info: www.lib.montana.edu/responsible-...

#DLFForum #DLF2025

Example system prompt for a research agent Example system prompt for a research agent. GitHub Gist: instantly share code, notes, and snippets.

Reasoning with Small Language Models (SLM) to Create Trustworthy GenAI

Slides: doi.org/10.5281/zeno...
System Prompt: gist.github.com/jasonclark/1...
Model Context Protocol server:
gist.github.com/jasonclark/4...

#DLFForum #DLF2025

The Responsible AI in Libraries and Archives team has released Viewfinder: A toolkit for values-driven AI in libraries and archives that was created by librarians and tech ethicists at four universities.

Print-at-home PDF: osf.io/yue9s
Interactive website: www.lib.montana.edu/responsible-...

Design of Future Scholarly Communication | Proceedings of the 43rd ACM International Conference on Design of Communication https://dl.acm.org/doi/full/10.1145/3711670.3764636