From: Greg Kuperberg <greg@matching.math.ucdavis.edu>
Subject: Responses concerning math archives
To: dmd1@lehigh.edu
Date: Sat, 3 Jan 1998 22:20:03 -0800 (PST)

As a companion to the previous announcement, I wanted to discuss
the objections that have come from a few archive maintainers (namely
Clarence and one other who has declind so far), and I wanted to respond
to some relevant recent comments by Mark Hovey in this forum.  I think
that the objections are minor compared to the potential benefits of
merging archives.  I invite you to judge for yourself.

The most strenuous objection has been the restriction that you have
to send TeX source to xxx.  Before xxx had automatic TeX compilation,
they didn't impose this restriction; some of the older submissions are
DVI or PS.  Afterwards, it became clear that the auto-compilation system
was robust and led to fewer failed retrievals than user-submitted DVI
or PS.  In addition, if you read the e-prints with Acrobat or xhdvi,
the system hyperlinks equation numbers to equations and citations to
the bibliography.  Some people worry about plagiarism from TeX source
submission.  This just hasn't happened at xxx nor at non-xxx math
e-print archives.

Another question is whether authors really want their e-prints transferred
to xxx.  In practice, very few mind.  xxx's policy is that any e-print
moved from another archive will be immediately and completely removed
at the author's request.  One of the archive maintainers who did agree
to merge was especially concerned about permissions and copyrights;
he plans to poll his contributors for a trial period of a month to make
sure that no e-print is moved against the wishes of its author.

On September 3, Mark Hovey wrote:
>The job market is so tough it seems to me that we should do everything
>we can to help each other out.  There are a lot of us on this list,
>and collectively we should be able to do something.

In my opinion, the biggest problem facing algebraic topologists on the
job market is lack of visibility: they see each other but outsiders do
not see them.  Last year there were tons of letters of recommendation
that read:  "Never-heard-of-him is one of the best young people in the
area of fill-in-the-blank in the past few years..."  Unfortunately,
too many letters of recommendation, from good and bad candidates alike,
have the same guarded, formulaic superlatives.  If a faculty member
has actually heard of a mathematician before seeing his or her file,
it makes a huge difference in the crucial first cut when the search
committee makes a short list.

Many job-seekers in algebraic topology face the tragedy that their files
will only be considered by departments that already have algebraic
topologists, and then only if the algebraic topologists prevail over
the rest of the department.  The most successful candidates are those
whose work has connections with other areas of mathematics, such as
commutative algebra or geometric topology.  The xxx math archives are
therefore a big opportunity for algebraic topologists on the job market.
(Indeed I will tell CoYM that they're a big opportunity for every math
job applicant.)  Workers in any given area of mathematics act against
their own interests if they conclude that their own home-grown e-print
distribution system works well enough for them, and who cares about
outsiders in any other area.

Finally, Mark also advertised John Palmieri's perl script called "bibweb".
I think this script is a nifty idea and I'm going to add it to my list
of features to add to the Front.  If I do implement it, you will be able
to upload your .blg file to the Front with Netscape and the Front will
give you back missing bibtex entries for those e-prints available at xxx.
You will not need to install any software to use it and it will be based
on the Front's readable search syntax.

                                Greg