we're in trouble! - (2 points)

Don Gilbert gilbertd at sunflower.bio.indiana.edu
Tue Jul 5 14:24:24 EST 1994

In article <199407051407.HAA21644 at net.bio.net> SCHLOSSER at ciit.org (Paul Schlosser) writes:
>I'm not saying that journals shouldn't become electronic.  The possible
>benefits are enormous.  And if printed versions are completely eliminated,
>then the efficiency would go up quite a bit (cost per published page would
>drop).  But there will still be costs for personnel, maintaining the server
>on which the info. sits, etc.  The question is, how should these costs be
>covered?  One possibility is to have NIH/NSF to provide these information
>services, with the expenses coming out of their budgets (leaving less for
>grants).  Then there would not be a direct user fee and the service would be
>perceived as being "free".  Anyone in the US [the world?] could log onto the 
>server and fish around the files for as long as he/she likes.  This will 
>require plenty of connections, since, without a fee, lots of people will be 
>using it, so the cost to NIH and/or NSF won't be trivial.  Researchers will 
>compete for access with anyone and everyone.

IUBio Archive is close to, if not, an electronic journal, full of
electronic publications (mostly software and data now, research reports
will come later).   It has not been explicitly funded by any agency, though
a group of archive users has generously contributed funds for disk space,
and I've spent plenty of time on it (more than I'd ever thought). 
The current computer it runs on was purchased thru an NSF grant to 
the Drosophila stock center, which uses the computer for related 
purposes (FlyBase e-publication).

This archive currently is serving over 4,000 transactions PER DAY to
biologists around the world.   A few years ago, I ran the archive
on my desktop computer (a Macintosh), which handled the smaller load
well.  The primary need for an electronic-journal-server is disk space
on an Internet connected computer.  Though I'm trying to get grant funding
to beef up IUBio archive and improve its quality, one can do *lots* of
electronic publishing on a small budget. 

Given the great desire for and utility of information servers that allow
free access by any (general public, students, teachers as well as 
resesarchers), and the small basic costs for hardware, it makes a lot
of sense for goverments to fund electronic publication basics.  The
expensive part is the time of people who will organize information
on any such e-publication servers, and support for software developers
to design better ways to give the public access to biology information.
High schools, grade schools, teachers and the general public are now
using biology information from IUBio archive.  It is accessible to them
as the Internet starts to reach into our public library and school 
system .. perhaps more accessible than many biology paper journals
(I've just found a few bio paper journals I want articles from that
our library doesn't have... too expensive).  If all of you authors
of biology articles would start sending copies to IUBio archive[*], 
you could be sure to reach a larger potential 
audience than just having it printed in a paper journal.

>BTW, one important aspect of journal publication is the peer review process--
>when you find something in a reviewed journal you have some additional 
>confidence in it's accuracy.  To replace this we will need electronic
>services where only reviewed articles appear.  While NIH might insist that
>every grant recipient publish a report on the product of their research, and
>could provide "free" access to this, that can't *insist* that it pass peer-
>review since some efforts will not make that criteria.  Should NIH have a
>listing of reports without peer-review?

Peer review will need to be incorporated into e-publishing before it
is widely accepted.  But keep in mind that e-publishing can allow new
forms of peer review.  The current paper review system is really peer *preview*
in the sense that a small number of readers preview a paper and allow
or deny its publication.   In an electronic journal, it may make more sense
to publish every thing but weight each document by readers comments
and votes.  Then the most widely read documents, with the fewest negative
comments/votes, may appear at the head of a list of available documents.
Poor or uninteresting documents would be available to read, but would
be clearly indicated as less interesting (to the general reader).
This could reduce problems with editors not being able to find
people w/ enough time to preview documents, or with selecting a poor 
subset of the peer population to provide an accurate preview of the document.

-- Don

[*] Seriously, send any biology-related preprints or reprints or e-prints
via anonymous FTP to ftp.bio.indiana.edu:/Incoming/.   Please send
in Microsoft's RichTextFormat (RTF), though postscript &/or plain text
will also be useful.

-- d.gilbert--biocomputing--indiana u--bloomington--gilbertd at bio.indiana.edu

More information about the Bioforum mailing list