On Mon, 24 Apr 2000, Dr. John R. Skoyles wrote:
> ...for the informed, any paper in Nature will retain
> the quality control reputation associated with that journal, for those
> not so informed [students, lay persons], publication source will not
> matter -- with the result there will be a leveling down of reputation.
>
> One solution is to limit eprints to peer reviewed journals. But that
> requires a filter that knows what journals are peer reviewed and which are
> not.
You have answered your own question. Who/what "knows" what is and is not
peer-reviewed in paper? Let that same (nonexistent and unnecessary)
oracle perform that (non)function online too.
Again, you are thinking along the same lines as those who are concerned
with making the Net or TV Networks kiddie-safe. The primary reason for
freeing the research literature from paper and paper-access tolls is
for the sake of research and researchers, not students and the general
public. (And students, fortunately, still have some guidance and
constraints from their teachers on what sources to cite and rely on.)
> Are there automated quality controls that could differentiate the
> status of eprints?
The answer is yes: Once the entire full-text corpus is openly archived
on the Web, there will be ways to do software-agent-based content
cross-checks to ferret out all kinds of things, including plagiarism,
altered bootleg versions, and papers pretending to be refereed or
published when they are not.
Again, this will not be a substitute for peer review, but a supplement
to it.
> software developers aware of quality control issues in their own field
> -- and obviously possessed of the programming skills to do it -- have
> invented techniques to automate it.
The online implementation of peer review is already taking advantage of
all available technology -- for speeding up the process, and making it
more efficient and equitable -- as that technology develops. But at
bottom, there are a few invariants: The editor's judgment (and
misjudgement) in selecting the right referees to review a paper has not
yet been automated (once it is, I assure you I will use it: I have
better things to do with my time); the same is true for refereeing
itself (the day a programme will do it at least as well on their
behalf, referees will be happy to return to their labs); the same is
true for the editor, again, in formulating a disposition letter, based
on the paper and the referee reports; even the author is waiting
patiently for the programme that will automate his revisions to conform
to the disposition letter (perhaps one will eventually do his research
for him too).
But in the meanwhile, the current system must proceed apace, and there
is not yet any substitute for real-time expertise in sight.
> The science community are not the only ones interested in quality control.
> Software developers are as well.
So are egg-certifiers. But not all forms of quality control are of a
muchness. And usually before you can automate a skill, you need first to
understand it (and there's still a bit of cognitive science left to do,
yet, till we understand egg-grading, let alone paper-grading).
> Unlike scientists, however, they have the abilities to be in the
> vanguard of programming its internet automatization.
First they have to understand what "it" is.
> That they have done so, I suggest changes the debate about eprints -- it
> means eprint archives need not necessarily level down reputations.
The "levelling problem," as I said above, is a nonproblem. Nor is that
what the "debate" about eprints is about. (The debate has been about
other nonproblems, such as copyright and preservation.)
> Techniques exist that will enable papers in eprint archives to gain
> differential status.
And they are the same "techniques" they have always been: Get your paper
refereed, revised, and accepted by the highest quality journal. (A good
prerequisite is first to try to do the highest quality research.) The
journal brand name (and citations) will then confer the differential
status, as they always did. Open access on the Web will then enhance
their visibility and impact.
> That should kill one source of criticism against freeing the scientific
> literature from its bondage to paper, and its access tolls. This
> off-topic subject as a result could turn out to be very relevant.
In the contrary; this underinformed talk of new unvalidated forms of
"peer review" on the Web will just strengthen (groundless) reservations
about the new medium.
--------------------------------------------------------------------
Stevan Harnad harnad_at_cogsci.soton.ac.uk
Professor of Cognitive Science harnad_at_princeton.edu
Department of Electronics and phone: +44 23-80 592-582
Computer Science fax: +44 23-80 592-865
University of Southampton
http://www.ecs.soton.ac.uk/~harnad/
Highfield, Southampton
http://www.princeton.edu/~harnad/
SO17 1BJ UNITED KINGDOM
Received on Mon Jan 24 2000 - 19:17:43 GMT