[EP-underground] The mudtrack ahead

So what next for eprints?

I'm not going to do much development work for a couple of weeks while
I wait to see if there are any major bugs that need patching.

I've got a few ideas on the next features to add, comments welcome:

 * Subject Editors - The admin can store a search, much like the subscription
      stored search, as part of the record for an editor. This editor will only
      be able to edit things which match this search. Useful for larger sites
      with several editors covering seperate areas.

 * HTTPS support - Security is good. Passwords over HTTP are "iffy". It's not
      life and death to us, but it may be to some people. The plan is to add an
      (optional) configuration option which will cause the parts of the system 
      which require a password to be serverd from an HTTPS server.

 * EPrints 1 importer. Now eprints2 has subscriptions this can be properly done
      (ish). It won't be automatic but will automate much of the process.

 * New option for subscriptions to allow you to select if you want to not
      be mailed empty subscriptions. My daily no-new-matches mail is getting
      really annoying. But if I had a monthly subscription it'd be handy to
      know that I was still subscribed.

 * Editor subscriptions - probably tied into the "Subject Editors" feature -
      this would mail editors when there's stuff to edit (handy when you do
      get frequent submissions)

 * BibTex (or other format) export. This could be handy if you use bibtex as
      part of latex. Most people probably won't care, but those that do will
      really care.

 * BibTex (or other metadata format) importer. This would be a big "textarea"
      box which you cut'n'paste a bibtex record into and it does it's best to
      start filling out the metadata for a record based on that.

 * Improved name searches. Currently we don't ignore umlauts in name searches.

 * "Offline indexing". Currently when records are added the system indexes 
      every "important looking" word in the metadata. And worse, when they are
      edited it deindexes THEN indexes. This is how when you search for "frog"
      the system can (almost) instantly tell you all the papers with frog in
      the title, as it made a note earlier in case anyone ever wanted to know.
      But the indexing is SLOOOOW. A better solution would be to run the 
      index/deindexing/reindexing etc. via a seperate script (once an hour, 
      say) which would speed up the user interface.
       
-- 

 Christopher Gutteridge                   [email protected]
 ePrints2 Coder, Support and Stuff        +44 23 8059 4833