Re: [tlug] Open Access Journals

["Stephen J. Turnbull" <stephen@example.com>]

Raymond Wan writes:

 > I'm more worried that the "open" in open access journals is not
 > quite the same as open source software.

I don't understand what this means.  Of course it's not; even if
copyright were abolished, academic professional ethics forbids
"reusing" somebody else's work as though it were your own.  So what do
you mean?  Just "who pays?" is different?  But is it?

 > Open access journals have solved that problem by charging at fee up
 > front.

But not all of them do.

 > Well, one "good" thing that publishers still do is quality control.

Publishers can't.  Editors do that, including making decisions about
retractions and the like, as well as recruiting competent referees.
But I suggest below that editorial boards can easily defect to
electronic journals.

 > And if something ends up being wrong, they'll hopefully act quickly
 > to retract the work.

But that's not the way it works.  Merely "wrong" articles are never
retracted, nor should they be.  They should be debunked and never
cited after the debunking (except in textbooks explaining how not do
research!)

It's fraudulent articles that get retracted.  For example, the "STAP
cells" paper -- if the research had merely failed reproduction in
other labs, there would be no talk about withdrawing the article.
Note that the original "cold fusion" article was never retracted
(although several supporting articles which were based on carelessly
conducted procedures were withdrawn -- mostly after being accused of
worse things than "carelessness").

The main technical problem with the "STAP cells" paper is that the
research protocol wasn't properly described, which is all too
common.[1]  But the search for hints as to the actual protocol led to
close examination of the article, which displayed many forms of
academic dishonesty according to reports.  Perhaps some of the
coauthors were honest, but it's rather hard to imagine that Dr.
Obokata was unaware that the project as a whole was at best taking a
lot of inexplicable shortcuts (eg, photographs of obsolete equipment
never used in Japanese laboratories).

 > With the web, that's difficult to do.

Actually, print is much harder to retract effectively.  I suppose they
could hire the "diary slasher".

 > Also, as long as people's careers depend on where work is published,
 > journals (and conferences) will probably have a place in academia.  I
 > can't see how the web could take over this role completely in the near
 > future...

Conferences, no, you can't replace them for the foreseeable future
(ie, as long as we can predict the orbit of the Earth around the Sun).
There are some things you can do at conferences that you can't do
between a keyboard and a chair.  But the academic publishing industry
could easily be replaced by a (properly organized) subset of the web,
and probably will be in the not too distant future in many fields by
well-respected editorial staffs defecting from their (unpaid)
positions with print journals in favor of providing more efficient
dissemination.



Footnotes: 
[1]  A careful study of economics articles using publicly available
data done in the 1980s found that only 50% could be replicated using
only the published article -- there was no issue of difficult manual
techniques as in cold fusion or STAP cells, it was purely an issue of
unreported statistical variables or "filtering" of the data.