Re: ACS meeting comments on e-prints

From: Albert Henderson <NobleStation_at_COMPUSERVE.COM>
Date: Wed, 5 Apr 2000 20:50:32 -0400

on 4/5/00 David Goodman <dgoodman_at_Princeton.EDU> wrote:

> It is not the case that Medline or its precursors ever covered or attempted to
> cover the "entire field of biomedicine." No one index does that or ever did;
> my idea of a really comprehensive approach I know of is: Medline + Excerpta
> Medica + Biosis Previews + Zoological Record + Agricola + CAB + Chemical
> Abstracts + Science Citation Index (and undoubtedly other indexes I don't
> regularly use). Some of these are government; some non-profit; some
> commercial.

Argue with my source, Mary E Corning and Martin M Cummings MD
(assistant director and director, respectively, National Library
of Medicine) Biomedical Communications, in ADVANCES IN AMERICAN
MEDICINE, edited by J. Z. Bowers and E. F. Purcell. New York:
Josiah Macy Jr. Foundation. 1976 p. 722ff:

        "The 20,169 articles in 1879 represented essentially
        the totality of the biomedical literature, whereas
        the 217,485 in 1973 were published in only 12 percent
        of the world's biomedical literature."

The tenfold increase in citations in 100 years is pitifully far
behind what Derek de Solla Price called the "modern normal rate"
-- doubling every 15 years or around 100 times in 100 years.
(Science Since Babylon. revised 1975. New Haven: Yale U. P. pp.

Let me cite similar analyses:

        The National Library of Medicine survey of periodicals
        received during a three month period estimated that the
        total number of substantive articles is roughly double
        the number actually covered by the Current List of Medical
        Literature, then the largest medical index in the world.
        (Brodman, E, and Taine, S I. Current medical literature:
        A quantitative survey of articles and journals.
        Proceedings of the International Conference on Scientific
        Information. Washington, D.C., 1958. National Academy of
        Sciences-National Research Council, 1959. 435-447)

        The biomedical literature that is accessible through
        electronic searching of bibliographies today represents a
        fraction of all studies in a field.(Scherer, R W., Dickersin,
        K and Kaplan, E. in Editing the Refereed Scientific Journal:
        Practical, Political, and Ethical Issues, edited by R A.
        Weeks and D L. Kinser. New York: IEEE Press. 1994.p 120-125)

Incidentally this deficiency presents a particular problem for authors
inasmuch as standards followed by many publishers ask researchers to,
"state general interpretation of the data in light of the totality of
the available evidence." (Begg, C., M. Cho, Eastwood, S. Et al. J A M A
276:637-639. 1996) It also implies referees must fly by the seat of
their pants.

Heaping praise on a mediocre excuse for an information service
does no service to the biomedical community, consumers of
health care, and sponsors of research.


> Albert Henderson wrote:
> >
> > [snip]
> >
> > Floyd E Bloom quoted a Senate advisory panel's
> > observation from 40 years ago: "The case for a
> > Government-operated, highly centralized type of
> > center can be no better defended for scientific
> > information services than it could be for
> > automobile agencies, delicatessens, or barber
> > shops." (Science 285:197 1999)
> >
> > The point is that scientists exchange their
> > work for dissemination and recognition by
> > their peers. It is best done in the private
> > sector. Government management always looses
> > any hint of excellence, particularly in the
> > information area. Look, for instance, at
> > Index Medicus. Coverage originally took in the
> > entire field of biomedicine. Now it covers about
> > 10 percent according to its own estimates. Or
> > look at the stagnating database used by the
> > National Science Board to create tables for
> > in the world it is supposed to indicate. Or see
> > how the government has run NTIS into the ground.
> >


Albert Henderson

> .
Received on Mon Jan 24 2000 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:44 GMT