I’ve been at the Royal Society conference on the future of scientific publishing at the start of this week, and I had the immense pleasure to chair the panel on infrastructure. I want to note my very first impression of the meeting. I have north of 20 pages of notes to work through, and some really interesting recommendations and suggestions for how to improve our scholarly communication system.
Last December I was at a HHMI invited meeting on this topic, and two weeks ago I was at the Metasciecne conference.
I want to sketch out some of the differences that I felt about these three meetings.
HHMI meeting
This was an intense workshop like environment that was very solutions focussed, with mostly open science, open infrastructure participates, along with some working academics, and hardly any publishers. Maybe something like 50 people in the rooom. It was highly interactive, really interesting, moderately optimistic, though with a slight lack of clarity on what the collective answer was that we were trying to get to. I like the approaches that were mooted, but felt that they are appropriate for a small audience, and might struggle to gain more widespread adoption. Of note since that meeting one of the entities present has had it’s funding collapse in a very public way (PubPub) and another has raised VC money (Curvenote).
Metascience 2025
Something like 400 to 600 people in attendance, mostly funders and academics, with very few librarians, and very very few publishers. I felt that this was a very optimistic conference, it was sort of brining a few slightly disparate fields together - science of science people, metascience people - bibliometricans. It felt a lot like a classic academic conference with many presentations of work in progress, but lots of examples of experiments happening in how to fund, and I felt a real belief in the power of funding to effect change, and a real belief that we could make positive change to our current funding mechanisms to improve outcomes, and the research ecosystem.
Future of Scientific Publishing Conference.
A two day meeting with many of the great and the good of scholarly communication, heavy on panel presentations, a forum for the vast and various accounting of the worries, grievances, and concerns, that have long stalked publishing since the advent of commercial entities in the ecosystems. Mainly present were librarians, publishes, some funders and government folk, and a small but not unimportant representation of actual researchers. It definitely felt the most pessimistic of all of the three meetings, and it felt like many of the questions we have been wrestling with for the past twenty years are still there. I think there are a few things that have moved on. We are in the post Open Access phase where the real complexity of that movement is now apparent. A thing that wasn’t mentioned was that we are in a hyper consolodative period for commercial publishers. Finally there are norms that are now expected (FAIR, CREDIT, deep use of DOIs) that are not controversial, but are somewhat patchy in implementation. I think what was good about this meeting is that it brought many of existing frictions explicitly to the surface in a very respectful forum for conversation, and if you are going to policy build it is important to do so from where we are, rather than where we might hope to be. There are trends that are important to me that didn’t get as much airtime as I would have given them, but equally there were important topics that I would have naturally skipped over that I got a lot of value out of.
I have much to think about and given time much to write about.
Some points that I thought I heard (no context, and not verified)
- Publishing is 1% to 2% of the cost of research.
- Our current system is as it is now due to the connection to the university, this was not the case pre 1940s.
- There are 70K journals (100K was later quoted to me in conversation later).
- Sieleo processes articles at a cost of $400.
- Elsevier fielded 3.5M submissions, published 700K and saw +600K increase in submissions year on year.
- UK institutes are considering sacking up to 30% of the roles in non research positions (not mentioned but this is being driven by the collapse of foreign student numbers).
- Early career researchers often don’t know who will pay for publication when they move labs - an issue I’d not been aware of before this meeting.
- English represents 68% of the research output (only).
- No one mentioned how DiamondOA can be funded sustainably.
- Mathematics has low impact factors because citations come so much later because it takes a long time to review a paper, every review is essentially a reproduction.
- While the market place of publishing is broken from a lack of pricing sensitivity, it’s not clear that OA has actually solved this problem.