FTP sites - directory & mirroring Charles Wells asked earlier whether anyone has a list of ftp sites containing category theory papers, and offered to maintain one. It is possible to do better than this, and "mirror" the files, ie copy them automatically. Lee McLoughlin, who maintains an archive of general software in my department, SRC.doc.ic.ac.uk has written a program to do this, available as /packages/mirror/mirror.shar from there. [Please note that this is NOT the same archive as contains papers by my colleagues in the Theory & Formal Methods section and my TeX macros: that is THEORY.doc.ic.ac.uk] All that is needed is to collect the relevant information in an appropriate form. I suggest that this is also a good place to keep address information in a machine readable format. My entry in this database reads: package=TaylorP comment=Papers by Paul Taylor at Imperial site=theory.doc.ic.ac.uk remote_dir=/theory/papers/Taylor local_dir=TaylorP # email=<pt@doc.ic.ac.uk> # phone=+44 71 589 5111 x 5057 # fax=+44 71 581 8024 # address=Dept of Computing, Imperial College, London SW7 2BZ, UK This is to be found in the file /theory/papers/mirror-file in the theory archive. This file forms part of the configuration data needed by mirror; other sites should provide similar files. It would be possible, I suppose, for me to extract this information for other sites by searching my "categories" and "types" mail from past years and exploring the various ftp archives, but I consider this is something that sites should do for themselves. So, if you use an ftp archive for your papers, please compile an entry in the above format, and put it together with the other files for the users of the same archive in an appropriate place. Then mail me either the file itself or a pointer to it in the archive. I will then keep these files in the directory /theory/FTP-sites on my archive, for general use, as well as using them to mirror your files on a weekly basis. B U T This will only work if people are organised about the maintainance of their archives. In particular, there should be one directory(-tree) per person. Don't mix files by several authors in the same directory. This will be a very useful resource if we just put in a little bit of work to set it up. Paul Taylor ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++