The Faculty of 1000 (F1000), in London, has announced an experiment in online science publishing, aimed at sharing research results widely and rapidly, and using open peer review to check postings afterwards.
The F1000 Research project, which begins publishing later this year and covers biology and medicine, will accept any format of work (examples given include posters, data tables, discursive speculation based on preliminary results, raw data sets and protocols) after an ‘initial sanity check’. It encourages authors to keep revising and updating what they have published. And by default, it will use open publishing licenses, allowing others to share and remix posted research (with attribution).
To reassure those worried that posting up data sets might nix chances of later more formal publication, there’s a long list of journals and publishers who say they wouldn’t view publication of data sets as ‘prior’ publication that prevents them from considering a more formal article.
The blog Retraction Watch characterizes the effort as similar to the well-established ArXiv site, which covers maths and some physics, and which marked its 20th anniversary in August. F1000Research isn’t quite the same as ArXiv: for one thing, it will charge fees to submit a publication, whereas ArXiv preprints are free. On its twitter feed (@F1000Research), F1000 says it is still working out the costs to authors for submissions, and will experiment with test papers over the next few months. Another open question is how much formal post-publication refereeing will be required — the project plans (unlike ArXiv) to invite referees to review published work. The biggest question of all is whether the project can get community buy-in; previous efforts such as Nature Precedings have attracted a few thousand submissions, although the emphasis there is on preprint manuscripts and presentations, not raw data sets.
Retraction Watch wonders how the new project would deal with work that needed retracting. Physicist Paul Ginsparg at Cornell University in Ithaca, New York, who founded ArXiv, e-mails me that ArXiv retains previous versions, but gives authors the possibility to retract. “Many authors do withdraw articles when they’ve found errors (particularly in mathematical proofs). Most responsible authors would rather not have something incorrect under their names, but of course it’s as low a barrier to withdrawal as submission to Arxiv was in the first place. Perhaps there’s more face to lose in withdrawing a journal publication,” he says.
As for coping with plagiarism, “there are some administrative removals (a few past plagiarism cases), and we’re prepared to freeze and publicly flag anything that is demonstrably ‘not compatible with Cornell academic standards’,” says Ginsparg. “But it has to be reported, and to be definitive without detective work since we don’t have staff to serve as an international policing body.”