Things to do
A rough list of things that it might be useful to add to the service.
I am not promising to actually do any of them anytime soon, but I'm
aware of them.
- Add more sites to the index (always)
- Truncate long entries, maybe to 1024 characters?
- Create a more effective method for browsing by site (done)
- Make the output sorted by time, with the most recent TRs listed
first (done, but maybe I should try to parse the date out of the entry
where possible instead of always trusting the time from
ls
-l...)
- Make a nicer format for sites with a consistent format (done for
bib; might someday be done for the WAIS-used %TI %AU etc. format;
parsing BIBTeX is a TeX-complete problem.)
- Allow users to use perl's power in regexps for more complex
keywords (done)
- Make it display the correct size of files even when they're
symbolic links (how?)
- Make a guide for TR archive site maintainers who wish to have
their work included here (done)
- Recognize addresses of people in entries and create anchors to
their finger entries(?)
This entire service is arguably just a hack; it's certainly not the
ultimate browsing mechanism for this kind of information. I hope
people find it useful, and that it may suggest ideas to others.
Ultimately, this should be done in a highly standardized fashion,
allow sophisticated queries, be WAIS/Z39.50ish in nature somehow,
interoperate with everything else, and be a supported network resource
someplace. Moving this service to that level is probably beyond my
ability and interest, but anyone wanting to do so may have (for
whatever they are worth) my code, my thoughts, and my blessing. I
hope this service will look like a clunky dinosaur a year from now
(assuming for the moment it doesn't already. :-)
(back to search
cover page, or to info on this
service)
Marc V