Thread by @realscientists

The view expressed below is part of what holds science back.

If you are privileged to have access to a "few colleagues" or be one of those colleagues, then the status quo works well for you. #PrePrint articles make this more seamless, & more equitable/

https://twitter.com/kanderson/status/1178292789571964929

The position of #PrePrint = not peer reviewed = rubbish, is also dated. We all know how hit and miss peer review can be. We also know how timely communication, with time stamping and dissemination, that is under the control of the authors (rather than a journal) is valuable.

For those of you who have not experienced peer review yet - we have a system in science where we ask our colleagues to review our publications. Our colleagues receive no money for this work (usually) and this is done all within confidential systems (managed by the journals).

This system is often viewed as the "gold standard" of quality control - it's not (more later) - as we are checking whether things are reasonable or make sense. We know that this is a flawed system, and are some very big case studies (as the book plastic fantastic explores).

The peer review process is managed by the journals - your paper will be handled by an editor (who may be an academic scientist), who will check a few things and then will send it out to peer review. The peers review the paper and advise the editor.

If the reviews are favourable, you'll usually be asked to make a few minor changes. Alternatively, the paper could be rejected outright or you could be given major revisions (e.g. do more experiments or analysis).

This peer review process can take lots of time, even before the paper is accepted. Usually we expect it to take ~6 months from submission of a paper until it is out, but it's a distribution (some faster, some much slower - years).

Now once the journal has your paper, and lets imagine it's accepted. They'll typeset the paper to their guidelines (even if you sent a typeset LaTex document) and a few other bits of work (like taking your copyright). Then they'll put it online, for others to read...

... they can read your peer reviewed paper if money changes hands, but none of this money goes to the authors, or the peer reviewers. It goes to the back office of the journals, maybe the editor, and profits to the shareholders (or maybe the learned society).

This money is paid in three ways - per article access (maybe £30/$30USD per 4 page article, I read about 20 of these a week), subscription (£ms, paid for by your employer), or the authors can select to make the article 'Gold' Open Access.

Gold Open Access means the article is free for everyone to read, but it will cost the authors anything from £250 to ~£7k for the article.

Now, we in the UK are unhappy with this. Our funders have paid for the research to be done, and they don't often want to have for it twice (or more than twice), so quite a long time ago we had a rule that items had to be made "green" open access, via an institutional repository.

We have an "act on acceptance" rule - within three months of the article being submitted, we have to put it online via our institution (with compliant embargo rules) - so that it's available for others to read.

A good institutional repository will be indexed by google - so you can get these free papers relatively easily.

Why does this matter for #PrePrints?

Well #PrePrints make the 'green open access' version available, straight away. It takes about 24 hrs for the a paper submitted to the ArXiv to be available online. The difference here is that the paper has *yet* to be peer reviewed, and it may contain some errors...

... but as I say before, the peer review process is fickle, and the accepted version may contain some errors. So the #PrePrint is a free version, available (and indexed by google scholar) and put up early. It can also be changed (eg if a random peer emails the authors).

The challenge here is that some people see the rise in #PrePrints as an erosion of the 'excellence' in research. However, preprints have been available for ~20 years. It's pretty much how all of the high energy physics work (i.e. particle smashing) gets done.

Now so far, some people suggest that "peer reviewed" journal publication is >> a #preprint. BUT some authors will get rejected by one journal, for flawed work, and just push it to the next journal (often without changes), until somewhere accepts it.

Do you know what is the gold standard of academic work? It's not peer review.

It's the rigorous error checking, re-running of experiments, and building on the work of others (shoulders of giants etc.).

This is made faster, and more equitable, by the #PrePrint culture.