the Peer Community journal [not the PCI journal!]

After a debate within the PCI (Peer Community in) communities as to whether or not create a Peer Community Journal, the decision was made to launch it and it is now ready. A really great initiative that I support (albeit from a fringe field when considering the range of the PCIs) and which can only succeed (to bypass traditional journals) if submissions happen. (There already exists a PCI Journal, specialising in research on precast, prestressed concrete. Not to be confused with!) To quote from the journal webpage,

Once an article has been recommended by a PCI, the authors can opt to leave it on a preprint server, to publish it in Peer Community Journal, to submit it to a PCI-friendly journal or to any other journal.

Peer Community Journal is run by researchers for researchers and is funded by public research institutions. It is:

• Unique = it is a single journal for all PCIs, and a generalist journal (Ecology, Evolutionary biology, Genomics, Archaeology, Paleontology, Network Science, Zoology, Infection, etc. )
• Free = it is a diamond open-access journal (free for both authors and readers), financed by public research institutions and Plan S compatible.
• Exclusive = it publishes only articles recommended by PCI
• Unconditional = it can publish any PCI-recommended article in its recommended version
• Opt-in = it publishes articles only if the authors wish it
• Immediate = after recommendation by a PCI, no delay between transfer to the journal and publication
• Community-based, with more than 1500 recommenders playing the role of Editors

This journal, created and funded by the PCI organization, accepts PCI-recommended articles without further peer reviews.

ten computer codes that transformed science

In a “Feature” article of 21 January 2021, Nature goes over a poll on “software tools that have had a big impact on the world of science”. Among those,

the Fortran compiler (1957), which is one of the first symbolic languages, developed by IBM. This is the first computer language I learned (in 1982) and one of the two (with SAS) I ever coded on punch cards for the massive computers of INSEE. I quickly and enthusiastically switched to Pascal (and the Apple IIe) the year after and despite an attempt at moving to C, I alas kept the Pascal programming style in my subsequent C codes (until I gave up in the early 2000’s!). Moving to R full time, even though I had been using Splus since a Unix version was produced. Interestingly, a later survey of Nature readers put R at the top of the list of what should have been included!, incidentally including Monte Carlo algorithms into the list (and I did not vote in that poll!),

the fast Fourier transform (1965), co-introduced by John Tukey, but which I never ever used (or at least knowingly!),

arXiv (1991), which was started as an emailed preprint list by Paul Ginsparg at Los Alamos, getting the current name by 1998, and where I only started publishing (or arXiving) in 2007, perhaps because it then sounded difficult to submit a preprint there, perhaps because having a worldwide preprint server sounded more like bother (esp. since we had then to publish our preprints on the local servers) than revolution, perhaps because of a vague worry of being overtaken by others… Anyway, I now see arXiv as the primary outlet for publishing papers, with the possible added features of arXiv-backed journals and Peer Community validations,

the IPython Notebook (2011), by Fernando Pérez, which started by 259 lines of Python code, and turned into Jupyter in 2014. I know nothing about this, but I can relate to the relevance of the project when thinking about Rmarkdown, which I find more and more to be a great way to work on collaborative projects and to teach. And for producing reproducible research. (I do remember writing once a paper in Sweave, but not which one…!)

news from PCI

PCI Math Comp Biol gets live!

A new Peer Community (PCI) preprint and postprint server is about to get live, with Mathematical & Computational Biology as its core interest. Thanks to the efforts of Amaury Lambert, Céline Scornavacca, and Eric Tannier. Following the earlier PCI Evol Biol (and my aborted attempt to start a PCI Comput Stats…). Although the funding and the core team are mostly French, the target is obviously international and editors from all backgrounds and specialties are most welcome to join as authors and reviewers!

open reviews

When looking at a question on X validated, on the expected Metropolis-Hastings ratio being one (not all the time!), I was somewhat bemused at the OP linking to an anonymised paper under review for ICLR, as I thought this was breaching standard confidentiality rules for reviews. Digging a wee bit deeper, I realised this was a paper from the previous ICLR conference, already published both on arXiv and in the 2018 conference proceedings, and that ICLR was actually resorting to an open review policy where both papers and reviews were available and even better where anyone could comment on the paper while it was under review. And after. Which I think is a great idea, the worst possible situation being a poor paper remaining un-discussed. While I am not a big fan of the brutalist approach of many machine-learning conferences, where the restrictive format of both submissions and reviews is essentially preventing in-depth reviews, this feature should be added to statistics journal webpages (until PCIs become the norm).