I started off not liking this post at all, but then warmed to it towards the end: https://retractionwatch.com/2024/10/21/reflecting-on-research-misconduct-whats-next-for-the-watcher-community/ I think the main gist is that science may operate in a way that is different from the way some people think it operates and if that is the case,...

I like the authors of this piece a lot, but the piece itself doesn’t say very much at all. https://blogs.lse.ac.uk/impactofsocialsciences/2024/11/13/preprints-at-a-crossroads-are-we-compromising-openness-for-credibility/. It does make me think that preprint servers should be the crucible for experimenting with every possible type of au...

https://katinamagazine.org/content/article/open-knowledge/2024/what-have-we-learned-from-subscribe-to-open I was not sure what to make of the above piece about subscribe to open, it’s not a model that I know very well. So I asked GPT O1 to do a critical read of this piece and the below seems fair. : From the evidence in the text, Subsc...

As we get to the end of 2025 I want to link to two reports that I was interviewed for over the last year. The first looks at future health trends, looking out to 2035. https://www.thelinusgroup.com/health-2035 The "Health 2035" report shifts focus from technology-driven solutions to the perspectives of young physicians, the future lead...

If you work in scholarly publishing, in any capacity really, I think you should go and read the recent report from STM on trust and identity - https://stm-assoc.org/new-stm-report-trusted-identity-in-academic-publishing/. I have been thinking about the role of identity, and identity verification, in our industry for a few weeks now, an...

https://www.knowledgegategroup.com/#platform won the Vesalius innovation - https://karger.com/pages/vesalius-innovation-award. They create a network to connect experts with those who need access to that expertise. I love this idea. I've worked on a few ideas like this in the past - Nature Network had some elements to it, we kicked some...

HumanLayer API and SDK for integrating human oversight into AI workflows, enabling approvals and feedback via platforms like Slack and email. LangChain Framework for building applications with large language models, offering tools for reasoning, data integration, and context-aware agents. ChainForge Open-source tool for prompt engineer...

Some colleagues have been looking at research integrity tools . There is a partial list that they have put together https://gist.github.com/IanMulvany/e2478eb8723aaa7c77e4c9b607b6bf75. (this is not a list of tools that we are using) STM Integrity: https://www.stm-assoc.org/stm-integrity-hub/ PySciDetect: https://gricad-gitlab.univ-gren...

Microsoft just announced a publishing imprint - https://blogs.microsoft.com/blog/2024/11/18/8080-books-an-imprint-of-microsoft-launches-offering-thought-leadership-titles-spanning-technology-business-and-society/, and https://unlocked.microsoft.com/8080-books/. Stripe has had an imprint for some time now - https://press.stripe.com/. I ...

1) Hawking radiation is not what I thought it was - https://www.threads.net/@spaceaccordingtoskylar/post/DCdGRiiv5Iq?xmt=AQGzOpg_N0-0T9fDfXrx93xVU-uMRZ4-_UCPelTJV5APTw - rather than being mass lost to particle antiparticle pairs, it’s to do vacuum states being distorted in curved spacetime, which makes them look lint why are creating a...

Last Thursday was International Digital Preservation Day, I hope you all were greeting each other with "Happy Digital Preservation Day" salutations! While I myself didn't quite get there, I did participate in a really fun and informative webinar on the topic hosted by our friends at STM. The video is up now - https://www.youtube.com/wa...

You can see the call here - https://blog.joss.theoj.org/2024/10/call-for-editors. Joss is the Journal of Open Source Software, and the infrastructure runs entirely within GitHub. It's a great lightweight model where the tools support the mission incredibly well. They have a nice overview of their costs here - https://blog.joss.theoj.or...

AI Regular previous readers of this blog will know that I have a deep interest in how AI is, or is not, impacting our industry. Here are some interesting external links / tools that I think are worth checking out:Here’s a more readable version of the list with improved formatting: • Aider: aider.chat A CLI tool that interacts with Open...

I deeply disagree with this opinion piece: Privatizing Peer Review. Back in 2008, I wrote about the relationship between publishing and the patent system, and my views have not changed since then: Patents and Peer Review. For me, the key difference between these two systems lies in how they approach the issue of scale. I don't see a sy...

This post by Adam Day makes the strong case that detecting AI generated text does not solve the fraudulent paper problem because AI text generators are now useful for genuine research. https://clearskiesadam.medium.com/detecting-genai-beside-the-point-01f6c3d8e05c I agree with Adam on this point, and I want to expand on two further ide...

OpenAlex - https://openalex.org/ is an amazing open replacement for the Microsoft Academic Graph, and an open alternative to Google Scholar. They have been going now for about five years and are now putting together an advisory board. You can see their call for members for this board here - https://docs.google.com/document/d/1TVEsfqRLA...

This is a good fast scroll. https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/the-top-trends-in-tech The two things I found most interesting. 1. Two year demand for tech jobs slightly up, irrespective of recent mass layoffs in the US. 2. Investment in renewable energy dwarfs all other areas of future facing tech inves...

I spent a short amount of time this morning looking at tools that can visualise dependency trees in a small python project. I'm hitting a small bug with circular imports in a tiny toy project that I am working on. The recommendation in this blog post worked for me - https://medium.com/illumination/visualize-dependencies-between-python-...

I bought myself a kindle so that I can read books with less distraction, and broadly I believe that it’s working well. One of the books I am currently reading is the excellent Attention Span by Gloria Mark. Some notes on some really interesting ideas: • Anticipation of learning creates a reward mechanism, which browsing the web can tri...

Health sciences researchers: Reuse Dryad data and win up to $100k from the 2024 Dataworks! Prize The DataWorks! Prize, sponsored by the Federation of American Societies for Experimental Biology (FASEB) and the National Institutes of Health (NIH), recognizes the impactful role of data reuse on human health. Successful submissions must: ...

The UK government has a fund to support training around AI, you can read about it here - https://www.gov.uk/government/publications/flexible-ai-upskilling-fund/ai-upskilling-fund-application-guide#who-can-apply. This is who can apply, and for how much: Business sizeHeadcountAnnual TurnoverBalance Sheet TotalMaximum grant funding availa...

GenAI just over a year on I made some predictions in May of 2023 at the BMJ board offsite. Let's review where we are now compared to what I said just over a year ago. Distribution We said: LLMs could disrupt discovery services such as Google. Since then: Google have accelerated introducing LLMs into their own search results, this remai...

Some news worth catching up on: • Antropic launched Claude Sonnet 3.5 • Competes with Open AI’s GPT-4o on price and capability. • Important because it shows that OpenAI does not have a monopoly on high quality models. • Therefore we need to keep evaluating multiple models for our needs. • Copyright risk for use of LLMs for software has...

I think the video came out OK. I'm sure that my thinking is going to continue to evolve, but this is a fair snapshot circa mid 2024. https://bmj.wistia.com/medias/y4b9gfs9ae

There is a compelling case that can be made that in many cases data is more important any research paper that is attached to the data, and certainly at governmental and state level, access to underlying data sets more important than written reports. For scholarly publishing, the role of data, and what should constitute a first-class ci...

This blog post is outstanding - https://lethain.com/mental-model-for-how-to-use-llms-in-products/. I normally like to write a bit more when posting on this blog, but I can only highly recommend this read.

What is it that publishers actually do? The job we do is to provide platforms to support the dissemination of verifiable claims about the world, but the business of what we do is about marshalling attention, from getting the attention of authors for them to submit their work, the attention of researchers to review that work, and the at...

The ICO in the UK recently asked for a request for comment on GenAI and data privacy. With my colleague Timothy Morgan - BMJ’s Data Protection Compliance Manager - we drafted a response from BMJ to this request. The below is our response. Do you agree with the analysis presented in **this document ? ** It’s a clearly written analysis, ...

This interview in Vox paints a concerning picture about the future energy use of LLMs https://www.vox.com/climate/2024/3/28/24111721/ai-uses-a-lot-of-energy-experts-expect-it-to-double-in-just-a-few-years It points to this research: https://www.technologyreview.com/2023/12/01/1084189/making-an-image-with-generative-ai-uses-as-much-ener...

We walk lightly on this earth, and when you live in a city like London you are stepping through history every day. I was at the Young Victoria and Albert museum today (my first time there since the big revamp), and these shoes took me on somewhat of a journey. They were owned by Lupino Lane, a Hackney performer from famous performing f...