Claude released 3.7 a few weeks ago and along with it they released a command line programming assistant - Claude Code. This was in limited preview on launch but last week was opened up. I tried it out last night and this post describes the outcome. When LLMs came out I started testing by developing a python web app that could use sock...

Here are things on the web that I found particularly interesting or insightful in February. Computing inside an AI | Will Whitney by Will Whitney “The metaphors we use constrain the experiences that we build, and model-as-person is keeping us from exploring the full potential of large models.” I hope that this piece is actully somethin...

I had to put a slide together recently where I wanted to evidence some of consolidation activity taking place in our industry. We all know a lot has been going on. I took posts from https://www.ce-strategy.com/the-brief/ - the excellent industry newsletter from Clarke and Esposito, and I put them into Google's Notebook LM - https://not...

I’m interested in whether LLMs can assist in peer review, and in particular some of the technical aspects of peer review around things like statistical review. So every now and again I throw the following paper at an LLM - Mortality rates among patients successfully treated for hepatitis C in the era of interferon-free antivirals: popu...

Pre covid, pre family, I used to go to a lot of events, in particular there was a thriving set up meet-ups in London. that's all quieted down now, as far as I can tell, so how do you keep up to date with what is going on? I'm mostly using https://netnewswire.com/ as a feed reader, and I pipe in a ton of newsletters and blogs, both arou...

Wikipedia has a list of language models that gives some information about release date, training size, parameter size, you can see that list here - https://en.wikipedia.org/wiki/List_of_large_language_models. I took that data and did some human curation on it giving a rough "capability" level to some of the models - . I then asked GPT ...

Micro Review - 2024. • 1462km cycled in 2024, compared to 278km in 2023. • Turned 50. • Still climbing better than my 8 year old and my 12 year old, but their skiing is far better than my snowboarding. • Got three new teeth - implants, painful. • Ate freshly cooked artichoke hearts, a new culinary experience for me, I liked it! • Trips...

tags: #llms #ui #local-models #tool-review After following this post - https://simonwillison.net/2024/Dec/27/open-webui/#atom-everything) from Simon Willison, I installed Open WebUI - https://github.com/open-webui/open-webui) using this command: uvx —python 3.11 open-webui serve (As an aside, uv has just totally taken away all of the p...

https://paulkedrosky.com/four-ideas-sunday-edition-17/ • US problems continue, instability in government, and decline in reading standards. • US Tech has stronger dependence on international markets than any other US industry. • Indicators continue for shift in energy landscape. • Overall picture is one of uncertainty, but where there ...

https://paulkedrosky.com/four-ideas-sunday-edition-16/ • The set of short economic indicators at the top of the newsletter are all pointing negative. • US mobility and hiring rates are both very low, a bearish indicator. • Other indicators are around global instability. • One highlight - India is having increased power demands as its e...

A list of things I found interesting on the web in January. Where will AI be at the end of 2027? A bet by Gary Marcus “If there exist AI systems that can perform 8 of the 10 tasks below by the end of 2027, as determined by our panel of judges, Gary will donate $2,000 to a charity of Miles’ choice; if AI can do fewer than 8, Miles will ...

Research Scientist by Centre for open science. The project will elaborate a conceptual framework for how the research process is conducted and evaluated (e.g., study design, conduct, analysis, interpretation) and different researcher personas (e.g., peer review, replication, and discovery) to develop specific tasks that can be benchmar...

I am picking up a bit of steam around writing for my blog, so it’s a good moment to reflect on why I blog—or at least why I seem to be blogging at the moment. My reasons are not wildly different from anyone else’s who blogs, I imagine. 1. Help Me Think I find it useful to write about things in order to help me think about them. On that...

• More robust document references in Claude https://www.anthropic.com/news/introducing-citations-api • OpenAI can now drive a web browser (US only) - https://openai.com/index/introducing-operator/ • OpenAI is looking to lead an investment of $500B in infrastructure in the US (they should put that into NIH funding) • Chinese model R1 dr...

I’ll be taking part in a webinar for In Publishing on AI next Tuesday - if you are interested you can register here - https://register.gotowebinar.com/register/6293426879838177120. It's a great panel of experts, and the following will also be on the call: * Jennifer Schivas, CEO of 67 Bricks (Data monetisation) * Markus Karlsson, CEO a...

I got access to GPT tasks this week and decided to see if I could make a task. Unlike an RSS feed, or a Google Scholar alert, I'm assuming that this does a broad web search. Something like this could become a useful addition to workflows that needed regular updates, if one got to a place where one was happy with the black box that was ...

I had this amazing AirBnB in San Francisco once, it was back in the day, before it seemed like things might be falling apart. On the hill, it had floor to ceiling full fronted glass wall with a stunning view out over the Oakland bridge. It must have been 2012, 2013 or thereabouts. It was an open science meeting or one kind or another. ...

So my view on the January 12th edition is that it is telling us: • Economic movement and growth continues, irrespective of current climate crises in California, but food industry will be a point of pain here. • US is a significant outlier in terms of drug related deaths, reaffirming structural issues in the US. • Data science related j...

I had dinner last night with some good friends and over the course of many topics of discussion the current state of social media came up. I found the following articles insightful. If I have one take at all about all of this, I think I would say that when we talk about these systems it is incredibly hard to get to grips with their sca...

Addressing Research Integrity with Ringgold by CCC I’d somehow missed this— but CCC is creating a service to help identify researchers and their affiliations. This is a problem that is evergreen, in that I’ve seen various implementations going back a number of years, from biomedexperts (if I recall correctly, the founders went on to la...

#kedrosky #trends I subscribe to the weekly newsletter from Paul Ketrosky - https://paulkedrosky.com/). He has built a system to scan a lot of news sources and pull out what he classifies as important trends in the news over the previous seven days (he has a daily feed for paid subscribers). It’s possible to just look at them as static...

I liked this article by Oliver Burkeman - https://www.theguardian.com/lifeandstyle/2025/jan/04/asking-yourself-a-simple-question-can-transform-the-way-you-think-about-your-life-oliver-burkeman I like the idea that if I can ask the right questions my life can be better, do those questions act as incantations and could I just invoke that...

tags: #research #scholarly-publishing Dear reader, I have been blogging for a long time. More often than not, I am writing for myself, mostly my future self, but I am aware that just over 140 folk have subscribed to get updates for my blog, and I am grateful for your attention. I tend to decide to send email alerts for posts that are m...

Dec 2024 Interesting Links My approach to running a link blog by Simon Willison “Sharing interesting links with commentary is a low effort, high value way to contribute to internet life at large.” I get a lot of value out of flowing interesting links that people post, and this post by Simon inspired me to create a custom apple shortcut...

After getting injured on my first day of skiing, I had a bit more time on my hands than anticipated over the holidays. I noticed that this blog has an RSS feed, but it's only for recent items. I've also been blogging across many different platforms over many years, and didn't have an easy way to see all of the posts that I've written. ...

I drafted this post a good while back, and never got around to posting it. Now that we are heading towards the end of the year, I’m going through my old drafts. Some time ago I did some leadership coaching and in that worked through the role of feelings and how they can enhance or limit our views of ourselves. At that time the followin...

This preprint - Artificial Intelligence, Scientific Discovery, and Product Innovation* - https://aidantr.github.io/files/AI_innovation.pdf from Aidan Toner Rogers at MIT https://economics.mit.edu/people/phd-students/aidan-toner-rodgers indicates that AI may be having a beneficial impact in materials science research. The preprint looks...

Many folks have tried to build the article of the future (myself included), with the starting point of the research article. We know for sure that technology now supports integration of data, code, and narrative, and we know for certain that researchers work with a variety of complex tools, only to at the end distill them into the pape...

I finally sat down to read through the Ithaka report on GenAI and scholarly publishing. Even though the state of the art has moved on a lot since the interviews were conducted, I still highly recommend reading this report, it's very well written: https://sr.ithaka.org/publications/a-third-transformation/#post-321519-footnote-1 I used m...

I first heard this phrase at the start of November at the annual NEC publishing conference, by Fiona Romero a director at the Wikimedia foundation, in her talk - "Knowledge is Human: How Wikimedians are engaging with AI". I love this phrase, and I think it describes so well what scholarly publishing is here for. It's an idea that has b...