This is a start: wiki.icelist.is/index.php?ti...

Just when you thought it was safe to dive into MeSH terms... I'm popping up a bit late to give a peek at what I've found for new #2026MeSH descriptors. Behold, a #medlibs and adjacent #expertsearching thread!

Library land 📚 and genAI critics, I seek a long-form post that I neglected to save to my Zotero library.... It was a professor who was reacting to the inclusion of ProQuest's AI Research Assistant and some problems with the definitions it was offering for medical terms. Ring any bells?

So..., this year there are no #MeSH changes right? Only the title of the MeSH Browser website was updated to 2026? #medlibs #PubMed

Hey #medlibs & library folks - check your catalog for the Journal of the Academy of Public Health. You may want to reconsider that & refine teaching language for clarity about evidence-based health information this year. eagledawg.blogspot.com/2026/01/publ...

Nobody should trust AI. That doesn't mean it isn't useful... not trusting it is part of AI literacy.

Last week a CDC webpage about vaccine safety was edited. Using the "Changes" feature of the Wayback Machine we can see a "before and after" comparison: web.archive.org/web/diff/202... The Internet Archive's Wayback Machine is the change control system the web never had.

Your annual #medlibs Halloween reminder that "Chocolate"[Mesh] has only been in use since 2017. (OK, to be fair previous indexing is "Cacao"[Mesh] 1963-2016, but STILL.) Also, "Spirit Possession"[Mesh] is a thing, with only 28 results in PubMed, "Spirit Possession"[tiab] gets 144 results. 🎃🍫🍬👻

#medlibs #ExpertSearching #SystematicReviews

#medlibs #canmedlibs

#medlibs Me, as every library tech company introduces "AI-powered search" which does not work.

The government shutdown is going to affect the MEDLINE component of all the platforms that hold MEDLINE, right? #medlibs

Non-USA #medlibs might find this useful www.theguardian.com/us-news/2025...

More of this please #medlibs 📚

While I am not the best instructor, I do enjoy the questions I get when At FSCI2025 in the day 2 of my workshop on AI powered search, someone asked me (1)

[Blogged] The AI powered Library Search That Refused to Search - From Clarivate's Summon to Primo Research Assistant, content‑moderation layers meant mostly for chatbots are "quietly" blocking controversial topics from being searched aarontay.substack.com/p/the-ai-pow...

Ended up spending the rest of my day trying to figure this out - when using the combine query option to combine sets, #Scopus gives dramatically less results compared to when you just type it out as one long string yourself. Will probably send them an e-mail today! #medlibs

"Goodbye wasted reading time. Hello relevance." is one way Elsevier is marketing ScienceDirect AI. I dunno about anyone else, but I find the idea that the time spent reading scholarly literature is a waste depressing, dismissive, and infuriating. #medlibs www.elsevier.com/products/sci...

"An archive of videos that have been or we suspect will be removed from the officiaI NNLM YouTube." #medlibs 📚 youtube.com/@altnnlm?si=...

Things to know and do to prepare your library for a visit from law enforcement, including ICE agents. www.oif.ala.org/libraries-an...

‼️HOT OFF THE PRESS‼️ ⭐Updated Responsible AI use in Evidence Synthesis (RAISE) 3-paper collection on OSF: osf.io/fwaud/files/... ⭐ Now includes: 1️⃣Recommendations for practice for the main roles in the ecosystem 2️⃣ Guidance on building & evaluating AI tools 3️⃣Guidance on selecting & using AI tools

Editors and peer reviewers, have you ever wondered if a #SystematicReview meets standards of conduct and reporting? CEE now has a checklist ✅ to help rapidly assess whether authors have adhered to best practices. @biljana-macura.bsky.social @aspullin.bsky.social 🔗

How ‘correct’ should #evidencesynthesis be? How does this change if we add #AI? Help us understand expectations to inform best practice via this survey forms.office.com/e/ZwdicYQpjv Open until 2 Jul, ~35 mins. Part of @WellcomeTrust funded DESTinY @cochrane.org @wellcometrust.bsky.social

“Libraries are under attack,” said Brewster Kahle. At our Escaping the Memory Hole event, he made a passionate call to action for defending digital libraries amid cyberattacks, lawsuits, & censorship. 📚💿 🎞️📼🏛️ Full video ➡️ youtu.be/_L1VP8klL_g #MemoryHole #Preservation #InternetArchive

"Despite this dismal success rate, companies are going all-in on AI, driven largely by the belief that everyone else is doing it. fortune.com/2025/05/09/k...

This republican govt is casting its reactionary net wide: Flemish Unis are asked if projects include DEI ..whether they're committed to climate justice, have links w China, Russia, Iran or Cuba..They must confirm they protect women from “gender ideology” www.universityworldnews.com/post.php?sto...

The Trump administration fired one of the most respected, celebrated, and qualified librarians in America. The real reasons she was fired are genuinely terrifying. action.everylibrary.org/haydenaicong...

Library of Congress staff just showed what resistance looks like. After Trump abruptly fired Dr. Carla Hayden, the respected Librarian of Congress, he tried to install Todd Blanche, his former personal lawyer and current Deputy Attorney General, as acting librarian. Staff weren’t having it.

A new memorandum from the Dept of Defense listing specific subject headings to be used within military libraries to flag books with "divisive content." #LCSH #critcat 📚 media.defense.gov/2025/May/09/...

New blog post in which I: 📈 detail a new citations-for-sale scheme in Google Scholar 🧻 make the 395 fake papers generated through this scheme available as a dataset 💤 chastise Google for failing to maintain the most widely used academic search engine Enjoy! reeserichardson.blog/2025/05/06/g...

Hey #medlibs, any of you actively refusing AI summarizing functionalities into your searching platforms, even when they are free? Or are you all just accepting it, even when it sucks?

The solution proposed by generative AI boosters is to devote more and more class time to learning about their product, its uses, its limits (if any) — and less and less to the substance, the subject matter of the course. Solutions, problems — what’s the difference? None, as long as you’re using AI.

Gaaaahhh. Of interest to #medlibs ...so happy we pay so much to publishers for high quality work like this!

ERICA to the Rescue! Read about our volunteer efforts to rescue & revive ERIC, Dept of Ed's research repository. www.datarescueproject.org/erica-to-the... Thanks to crizzo, @archive.org & @safeguardingdata.bsky.social for all of the hard work. Check out more rescued data on our Data Rescue Tracker!

And, just like that, the top journal in the field of Environmental Health, published by the National Institute of Environmental Health Sciences, a part of the NIH, goes dark...

And here I must share (again!) my ongoing endeavour to ensure that #LibraryTwitter has continues in some form over here…!

4) I need some information and I wish to just pose my question to an all-knowing oracle. DO NOT DO THIS. Chatbots, even if they could reliably return "the" correct answer, are not a good tech for information access. buttondown.com/maiht3k/arch... >>

LLMs are nothing more than models of the distribution of the word forms in their training data, with weights modified by post-training to produce somewhat different distributions. Unless your use case requires a model of a distribution of word forms in text, indeed, they suck and aren't useful.

More than 130,000 government pages have gone dark in a purge that one scientist likened to a “digital book burning.” A group of librarians and archivists is fighting back.

Just an update: We have had some questions about ERIC. There are three access points in our Tracker (search ERIC), and we know of one more access point. Tracker: baserow.datarescueproject.org/public/grid/... SCIOP access: sciop.net/datasets/eri...

How did we get from "Do not use Wikipedia as a source" to "ask hallucinating chatbot everything"?

Our CEO @louiscoiffait-gunn.bsky.social highlights the “lack of robust evidence” about how widespread censorship is in the UK and which books are not available. www.theguardian.com/books/2025/a...

#medlibs and other #librarians #skybrarians

Dr. Bastian posted a link to the Literature Selection Technical Review Committee members, which now takes you to a page that says: Sorry, the committee list is not available at this time. Please check back shortly. We apologize for any inconvenience. internet.csr.nih.gov/Committees/M...

#MedLibs #CanMedLibs

NOAA just announced a long list of datasets that are going away🧪⚒️🌊: www.nesdis.noaa.gov/about/docume... Download what you need asap and send comments to: [email protected]

I can't emphasize enough that the most important thing journalists can do right now is publish exactly this kind of article

Es ist mittlerweile überall so: