I advise you all to read a set of four preprints I recently mentioned on the
SLA Physics Astron. Math. list which I will cite below.  In particular, look
at section 2.2  _Data Representation_ of CfA-4872, which explains why the
NASA Astrophysics Data System started reformatting all of their
bibliographic records as XML documents.  This particular paper is available
in various formats at http://xxx.lanl.gov/abs/astro-ph/0002105

CFA-4870
Kurtz, M.J., Eichhorn, G., Accomazzi, A., Grant, C., Murray, S.S., Watson,
J.M. (Harvard-Smithsonian Ctr. Astrophys.)
The NASA Astrophysics Data System: Overview.
Feb 2000
19 p.
To appear in A&A Suppl.
http://xxx.lanl.gov/abs/astro-ph/0002104

CFA-4871
Eichhorn, G., Kurtz, M.J., Accomazzi, A., Grant, C.S., Murray, S.S.
(Harvard-Smithsonian Ctr. Astrophys.)
The NASA Astrophysics Data System:  The Search engine and its user
interface.
Feb 2000
23 p.
To appear in A&A Suppl.
http://xxx.lanl.gov/abs/astro-ph/0002102

CFA-4872
Accomazzi, A., Eichhorn, G., Kurtz, M.J., Grant, C.S., Murray, S.S.
(Harvard-Smithsonian Ctr. Astrophys.)
The NASA Astrophysics Data System:  Architecture.
Feb 2000
25 p.
To appear in A&A Suppl.
http://xxx.lanl.gov/abs/astro-ph/0002105

CFA-4873
Grant, C., Accomazzi, A., Eichhorn, G., Kurtz, M.J., Murray, S.S.
(Harvard-Smithsonian Ctr. Astrophys.)
The NASA Astrophysics Data System: Data holdings.
Feb 2000
27 p.
To appear in A&A Suppl.
http://xxx.lanl.gov/abs/astro-ph/0002103

Rob Atkinson
Library
Fermi National Accelerator Laboratory
[log in to unmask]

----- Original Message -----
From: "Karen Coyle" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Wednesday, March 29, 2000 3:05 PM
Subject: Re: XMLMARC Software Released


At 09:23 AM 3/29/00 -0500, you wrote:

 What
>XML offers is the possibility, of having client applications that know
>nothing about MARC be able to format bibliographic records for display to
>the user.

Yes, the display possibilities of XML are exciting. I want to have the
ability to build a display from a variety of resources and databases, so
that my bibliographic information may come from a traditional library
catalog (that outputs XML) but on the screen that data will be combined
with full text, links to other resources about the author, etc., and yet,
at the time of display, this whole mess of stuff can be a single, logical,
XML document that allows for further navigation, retrieval, etc.

With XML it should be possible to have an information universe of parts
that can be recombined into a wide variety of wholes. I much prefer that
approach than one of trying to create a single system that has all of our
types of information stuffed into a single record format. We can store and
retrieve each kind of data using the data management systems best suited to
it, something that is key to creating large and efficient data stores. We
can afford to be less efficient, if necessary, after the major work of
retrieval is done.

And wouldn't it be great to present data to the user in a format that she
can independently manipulate with standard computer programs? (Of course,
the early promise that MicroSoft would use XML as its underlying format for
its word processing program has already fallen through. Undoubtedly they
didn't like not being able to tie up their data in a proprietary format.)

My hope is that future systems will take the "tinker toy" approach, each
one putting out data as interchangeable parts.

----------------------------------------------
Karen Coyle                    [log in to unmask]
  University of California Digital Library
  http://www.kcoyle.net        510/987-0567
----------------------------------------------