Keith Robison robison at
Mon Jan 15 20:14:47 EST 1996

U27111 at wrote:

: berezin at MCMAIL.CIS.MCMASTER.CA (Alexander Berezin) wrote on 12 Jan
: 1996 11:46:24 -0800:

: >That's what I am saying. Its pointless to attempt to
: >change the nature of people - it will remain the same
: >regardless. But you can modify practices in a given
: >field in such a way that miscounducting behaviour will
: >not bring the actor too much the avantage. In the above
: >example people tried to publish in Nature because
: >the present (arbitrary, fictional) reward system
: >gives a great premium to anything that published in
: >Nature (as opposed to elsewhere).

: >As I said before, I see it as a sheer stupidity
: >and anachronism (though deeply ingranted by
: >historical reasons). For me it makes no difference
: >at what paper (or by what publisher) my (or your)
: >article is published for as long it is published and
: >the community can find about its existence (say, thru
: >abstract services) and all who are interested can
: >obtain a copy of the full text.

: [playing at devil's advocate... asking the types of questions a
: typical Joe Lab would?]

: But then by what standards due we judge excellence in our
: community?

: Do we have to pour over countless papers looking for high caliber
: work... instead of supposedly having it available in a hand full of
: 'prestigious' journals?

: By track record? ...In biomedical research that could take years.

: And how would I know if something was worth calling my broker about
: if Science or Nature wasn't there to tell us which ones are
: 'mainstream' and *hot topics* - with full text to back it up?

: Just some thoughts,

Clever jokes aside, there are side-benefits from the current 
system which must be considered.  First, the centralization of
papers in journals does make it easier to find information.
Time spent finding information is time not spent doing something
else.  Perhaps abstracting services would be able to deal
with the sort of highly-decentralized publishing system which
Berezin proposes, but the current experience with the WWW
is cause to think twice about rushing into such a system.

Another important class of ancilliary benefits is the required
release of certain types of data -- in particular molecular
sequence and structure data.  Most journals (with two
prominent single-word-title exceptions [and Cell does]) 
require the deposition of sequence and structure data prior 
to publication; as a result much of this data is captured
that would otherwise be lost.  Such data is sometimes not
submitted out of forgetfulness, and sometimes for
competitive reasons.

It is interesting to speculate on the consequences of
a wholesale changeover to a no-review, no-journal publication
system.  I suspect a major effect would be an even stronger
interest in review journals, in order to provide centralized
points of reference to the even more dispersed literature.
This might, in a certain sense, constitute a form of
post-publication open review, and would be interesting
and probably positive.

Opponents of the current system would advance their case much
better (IMHO) if they led by example -- formed central collection
points for no-journal publications or by forming open-review
electronic journals.  Most of electronic science journals that
I know of are APR -- probably partly because it hasn't occured
to try another model, and partly because electronic journal publishers 
are worried enough about be accepted without tinkering with the
format.  Those who are true believers in NPR or OPR journals
are the people who believe in the concept, and would therefore
not be worried about trying it out.

Keith Robison
Harvard University
Department of Cellular and Developmental Biology
Department of Genetics / HHMI

robison at 

More information about the Bioforum mailing list