Dear colleagues, I'm writing to you about a proposed e-print archive in category theory, on behalf of a committee of mathematicians which is advocating the unification of various electronic preprint servers and archives into a single system with a common format. We want to invite Category Theorists to join this group. This invitation is being issued to 15 preprint servers (listed below) which together cover about half of the subject areas in mathematics. We are also trying to organize active groups in the remaining subject areas, so that the unified preprint server will soon cover all of mathematics. This organizational effort has already produced interested groups in combinatorics and geometric topology. Managing a preprint server can be a time-consuming task, but fortunately there is a similar preprint server in operation in physics, running at Los Alamos National Lab and funded by the National Science Foundation, whose staff is willing to take on the operation of a unified mathematics server. This will enable all of us to take advantage of their years of software development, as well as their efficient staff who monitor the (fully automated) operation. When scientific questions arise (such as the suitability of a particular paper for a particular archive), these staff members would consult with a designated "moderator" for each subject area. (The moderator also receives abstracts of papers as they are submitted, and can have the staff intervene in the rare instances of inappropriate papers or other difficulties.) Someone you designate collectively would serve in this capacity under the new system. The great advantage of a unified server is that all mathematicians will be able to participate in electronic preprint exchanges in a single, flexible system (which can distribute preprints in a variety of formats, including dvi, ps and pdf). The computer systems staff in a mathematics department can be asked to familiarize themselves with this system, which will be easy to support for users. Moreover, a central location for preprints (with dozens of mirror sites worldwide) together with a soon-to-be-familiar scheme for citing these preprints will mean that mathematicians in one field who see a citation to a preprint in another field will easily be able to locate it. We will also all gain the ability to simultaneously search through related archives, including full text searches. There is a common keyword index for the mathematical holdings, permitting simultaneous searching of all or designated related archives. A given paper could easily be relevant to users of more than one archive, so this would make information much easier to track. In order to take advantage of the system in place at Los Alamos Lab, we have agreed to follow the technical decisions they have made: authors are asked to submit tex source files, for example. Although this may be a change for users of your current system, we believe that the long-term advantages of unification outweigh the temporary disadvantages to users of an established archive. We propose to copy the contents of your archive (in its current form) into the unified archive, and suggest that you encourage all future submissions be made directly to the unified archive. Category theory is one of the subject areas in the unified archive and is intended as a continuation of your archive. It should be fairly easy to construct a kind of front-end on a web page at your site which will provide access to both old and new papers in your portion of the unified archive, giving your users a certain degree of continuity and allowing them to retain the feeling of belonging to a smaller community of users of a particular preprint archive. We will be happy to assist in the construction of such front-ends. A primitive prototype for such a front-end (in the area of combinatorics) can be viewed at http://eprints.math.duke.edu/archive/CO/ All but one of the links on that page take you directly into the archive at Los Alamos. A front-end for the entire unified archive is under construction at U.C. Davis by Greg Kuperberg: you can view it at http://front.math.ucdavis.edu You might also like to look at the Los Alamos Math Archive itself (partially functioning already), at http://xxx.lanl.gov/form/math You will see that the combinatorics and geometric topology sections are getting a good start this week. At the time of unification (probably Jan. 1) the four current math archives alg-geom, dg-ga, funct-an and q-alg at the Los Alamos site will be merged into the unified archive, becoming subject areas AG, DG, part of FA, and QA, respectively. Our committee has had Joe Christie, Greg Kuperberg, Dave Morrison, Dick Palais, Jim Stasheff and Mark Steinberger as its active participants, with occasional participation by a few others. If you are interested in having a representative of category theory joining this committee, we would love to have one. Our tasks after this unification will be developing the remaining subject areas, and publicizing this archive among mathematicians. Short of that, as individuals, all you need to do right now is write back and say you think the category theory category is a good idea, or better yet, to pledge some number of e-prints to be submitted soon after the category starts. Those already archived other than on your own server should be offered with a copy of the offer going to the archive administrator. Of course we are most eager to have your present and future e-prints added to the unified server. (For e-print versions of papers for which you no longer have the copyright, permission from the copyright holder will be needed.) I am very optimistic that this effort will eventually lead to a complete transition to efficient electronic distribution of new results in category theory. The xxx archives has 3,000 math e-prints in the four existing mathematics categories and 50,000 physics e-prints. As you might guess from these numbers, xxx has already effected the electronic transition in several areas of physics. The category theory category will be of great benefit to us all. Please help get it started; the more who join, the better. Best regards, Jim Stasheff P.S. Here is the list of 15 preprint servers which we propose to merge. If you have other suggestions for inclusion on this list, please let us know. 1. Algebraic Geometry (http://eprints.math.duke.edu/archive/alg-geom/) 2. Algebraic Number Theory Archives (http://www.math.uiuc.edu/Algebraic-Number-Theory/) 3A. Banach Spaces & Functional Analysis (ftp://ftp.math.okstate.edu/pub/banach/) 3B. Functional Analysis (http://xxx.lanl.gov/archive/funct-an/) 4. Combinatorial and Geometric Group Theory (MAGNUS) (http://zebra.sci.ccny.cuny.edu/web/html/magnus.html) 5. Conservation Laws Preprint Server (http://www.math.ntnu.no/conservation/) 6. Differential Geometry (http://www.msri.org/preprints/dg-ga.html) 7. Dynamical Systems Electronic Preprint Server (http://www.math.sunysb.edu/dynamics/preprints/preprints.html) 8. Hopf Topology Archive (http://hopf.math.purdue.edu/pub/hopf.html) 9. K-theory Preprint Archives (http://www.math.uiuc.edu/K-theory/) 10. Logic Eprints (http://www.math.ufl.edu/~logic/) 11. Mathematical Physics Preprint Archive at the University of Texas, Austin (http://www.ma.utexas.edu/mp_arc/mp_arc-home.html) 12. Quantum Algebra and Topology (http://eprints.math.duke.edu/archive/q-alg/) 13. Representations and Cohomology of Groups (http://www.math.uga.edu/~djb/archive.html) 14. Several Complex Variables (ftp://iu-math.math.indiana.edu/pub/scv/) ************************************************************ Until August 10, 1998, I am on leave from UNC and am at the University of Pennsylvania Jim Stasheff jds@math.upenn.edu 146 Woodland Dr Lansdale PA 19446 (215)822-6707 Jim Stasheff jds@math.unc.edu Math-UNC (919)-962-9607 Chapel Hill NC FAX:(919)-962-2568 27599-3250
I am all in favor of Jim Stasheff's proposal, but what does one do about the fact that I, for example, have my own personal macros files and my TeX files cannot be compiled without them. If this is to be a straitjacket, I would just as soon maintain our own server with dvi and ps files. I am also just a bit wary of distributing the tex source, since it is so easily changed. Not that dvi and ps files can't be changed, but it is certainly a good deal harder. Michael
you can input your macros directly into your tex file as for the plagarism issue, I will let the experts respond ************************************************************ Until August 10, 1998, I am on leave from UNC and am at the University of Pennsylvania Jim Stasheff jds@math.upenn.edu 146 Woodland Dr Lansdale PA 19446 (215)822-6707 Jim Stasheff jds@math.unc.edu Math-UNC (919)-962-9607 Chapel Hill NC FAX:(919)-962-2568 27599-3250
Michael Barr writes:
I am all in favor of Jim Stasheff's proposal, but what does one do about the fact that I, for example, have my own personal macros files and my TeX files cannot be compiled without them.
Preprint servers of the sort Jim is describing have facilities for uploading macro files, postscript files, etc. along with the TeX files. I use them a lot and they work. Best, John Baez
Jim Stasheff writes:
you can input your macros directly into your tex file
But you don't need to. You can also upload a bunch of separate files using the uufiles program obtainable from the preprint archive. It's painless and easy. Information on this and a million other technical issues is available at the archives themselves. Try: http://eprints.math.duke.edu/ or send email with subject header "help" to q-alg@eprints.math.duke.edu
as for the plagarism issue, I will let the experts respond
I'm no expert, but I don't see what the problem is. If you upload your paper to one of these archives, the paper itself and the exact date and time it was first received is publicly accessible, so any attempt by anyone to plagiarize it would be incredibly easy to prove. If you store your papers on your own site, it's much harder to prove you wrote them before someone else. This is one reason why highly competitive physicists rush to put their papers on the archives as quickly as possible: to get a certified time stamp on their paper! Anyway, I've never heard of any problems with plagiarism actually happening. In case it's not been made sufficiently clear, the main advantages of having all papers on a given subject stored electronically at a single institution are: 1) they are easy to find 2) they are easy to refer to 2) they stay there, archived, while the authors move from institution to institution and ultimately perish. Will our institutions keep our websites going after we die, while technology continues to change? Many of my papers (or pointers to them) appear on Ginsparg-style preprint archive, my own website, Hypatia, Mathematical Reviews, the category theory mailing list, and paper journals --- all of which serve different purposes. Presumably some of these systems will fall into disuse in a natural sort of way as time passes. I wouldn't advocate the brutal elimination of existing systems. I think the question now is: would category theory be served by creation of a Ginsparg-style preprint archive for the subject?
participants (3)
-
James Stasheff -
john baez -
Michael Barr