Re: First Monday article on OA

[Jan Velterop <velteropvonleyden@btinternet.com>]

Pre-publication and post-publication peer-review serve different purposes.

The most important role of pre-publication peer-review (PrePPR) is to
accept (or reject) an article on the basis of the scientific soundness of
the data and the arguments used; whether the data support the conclusions.
Often its role also implies accepting (or not) an article in the
scientific orthodoxy of the time, or, if it challenges the orthodoxy, pass
judgement on the coherence of the challenge. It can, but doesn't have to,
also imply judgements of quality and/or interest-worthiness.

Post-publication peer review (PostPPR) is mainly about quality and
interest-worthiness. Sound, worthwhile, but pedestrian articles may be
necessary bricks in the edifice of scientific discovery and understanding,
they are rarely perceived as interesting or quality articles to the degree
that they shouldn't be missed by researchers in the field or even beyond.

Articles that warrant PostPPR are almost invariably the ones that excite
the reviewer (as it is *their* choice, unlike with PrePPR, where they
rarely review articles of their own choice). They show up in review
articles, and in new, systematic PostPPR services such as Faculty of 1000,
which help the researcher stratify the articles in layers of
interest/quality/relevance in a dimension different from the usual
stratification that journals provide, on the basis of only their PrePPR.
Faculty of 1000 shows up exciting articles, many in relatively lowly
journals, and exposes, implicitly, the pedestrian nature of those articles
in high-profile journals that never make it to inclusion in PostPPR.
Moreover, it does so well before the number of citations to a given
article (usually lagging by at least a year) can give an indication of its
importance.

Having just PostPPR and no PrePPR may mean that many articles are not
reviewed at all. They may subsequently be perceived to be no more than the
glut of junk already present on the web. Hardly something we've all been
eagerly awaiting.

Jan Velterop

On 20 Aug 2004, at 00:13, Mark Funk wrote:

> I agree with David Groenwegen about the many potential problems with
> post-publication review for OA, which Joe Esposito espouses in his
> article at http://www.firstmonday.org/issues/issue9_8/esposito/ :
>
> "In a world of electronic networks, however, peer review can and should
> take place after publication. Such review can take place virtually in
> real time, so our patient remains safe; and the marginal cost of such
> publication (which essentially amounts to uploading the text to a Web
> server) approaches zero. Post-publication peer review does not require
> expensive and slow-moving infrastructure. It therefore assists authors
> in their goal of getting published quickly and potentially provides
> wider feedback from the broad community of scholars. Quackery that gets
> published will be recognized as such and dismissed rapidly."
>
> Why would post-publication peer review be any less expensive or
> slow-moving? Does moving the review process to after publication somehow
> make it cheaper and faster?
>
> Sure, you can have a *type* of post-publication peer review that is fast
> and cheap. Look at Amazon.com's customers reviews. It's fast, it's
> cheap, and it's woefully inadequate for science, particularly clinical
> medicine.
>
> "Real time" review to protect the patient? Current reviewers will often
> spend weeks going over the procedures and statistical methods of
> clinical research articles. (PLoS tries to have reviews done in 7 days,
> but realizes that sometimes more time is needed.) This is not "real
> time." Shortcuts to the review process for English literature may be OK,
> but I prefer my doctor to be practicing medicine that has been properly
> vetted, not waiting for review.
>
> I can't imagine any reputable Open Access journal actually putting
> articles on the web prior to review. It would only take a couple of
> articles exposed as quackery or fraud before that journal's reputation
> would be shot.
>
> Mark Funk
> mefunk@mail.med.cornell.edu