Re-sending as a result of mail host crash on 2/20.


---------- Forwarded message ----------
From: Jan Velterop <[log in to unmask]>
Date: Wed, 20 Feb 2013 19:52:51 +0000

Those who think publishers have a role to play with regard to
filtering the overwhelming amount of articles that are being submitted
and published are right, of course. Nobody can read it all. I've
earlier pointed to the article by Alan G Fraser and Frank D Dunstan
"On the impossibility of being expert" (BMJ 2010; 341 doi:
http://dx.doi.org/10.1136/bmj.c6815 — Published 14 December 2010) as a
good example of the situation.

Unfortunately, what I see and hear is that the role in filtering that
publishers undoubtedly have is invariably interpreted in terms of
deciding which article to publish and which to reject, with the help
of peer review and 'quality' criteria.

But almost every article submitted will be published, eventually,
after cascading down the journal pecking order. And no serious
scientist that I know of decides to read an article just because it's
published in a prestigious journal, or, perhaps even more to the
point, decides not to read it just because it's published in a lesser
journal. He or she tries to find must-read articles via searches,
consulting colleagues, following references, often without realising
or taking note of the journal in which they are published. Of course,
if the journal title is noticed, a scientist will interpret the
meta-information that the label of a journal attached to an article
conveys.  A paper with the label of a high-ranking title, one of the
scientific glam mags, say, may well be interpreted as "interesting,
quality, citable, but not repeatable or useful in the field" and one
with the label of a lesser journal as "practical, repeatable, useful
in the field, giving me ideas for my next grant application, perhaps
not too citable if I care about my career." (A friend, who worked on
tropical diseases, once said, only half-jokingly, that the Impact
Factor is inversely proportional to the actual usefulness of a
journal).

The really useful filtering, however, the way to use the knowledge
contained in all the articles that one cannot read for lack of time,
is to distill the scientifically significant assertions from the
waffle in the text and get the 'picture' in that way, by combining
large amounts of such assertions. A 'helicopter' view or hot air
balloon view (with, indeed, the noise or the hot air). An overall
picture of the lay of the knowledge landscape that enables you to
decide where to dig, what actually to spend time on reading. The role
publishers could play to enable this is to optimise articles for
machine reading, making them interoperable for data- and text-mining,
without the barriers that make that impossible. And to focus
copy-editing on the essentials for machine reading (my hobby horse,
because it goes wrong very, very often: correcting the German sz
ligature - ß - where clearly a bèta - β - is meant; they may look
alike, but they are very different to a computer). It's the only way
to stay relevant as publisher, I think. And the only publishing model
suitable is the 'gold', CC-BY, open access one, be it subsidised or
maintained by APCs. Not the 'green' model, delivering often little
more than 'ocular' access, solving yesterday's problems, not today's.

Jan Velterop