Re: Gnus and GroupLens

[davida@lsl.co.uk (David Allen)]

>>>>> "Brad" == Brad Miller <bmiller@cs.umn.edu> writes:

>>>>> "David" == David Allen <davida@lsl.co.uk> writes:
    David> Two problems:

    David> 1) When trying to enter a GroupLens enhanced group I get
    David> Symbol's function definition is void: make-local-hook I see
    David> this is also used in custom.el, but protected by (if
    David> (fboundp 'make-local-hook)) Presumably this function is not
    David> in XEmacs (yet).  Retrying seems to (usually) enter the
    David> group successfully.

    Brad> Hmmm, But I have my doubts about whether your ratings are
    Brad> actually getting sent to the BBB when you exit the
    Brad> group.....  You may want to use: (add-hook
    Brad> 'gnus-exit-group-hook 'bbb-put-ratings) until I fix this
    Brad> properly.

Well, it says "Sending Ratings...Done" when I quit the group.  I don't
know that I can check whether they are actually getting there.

    Brad> make-local-hook, is supported in xemacs-19.14

Roll on the day ...

    David> 2) Trying to enter the group the next day fails with
    David> timeout errors, presumably retrieving the GroupLens scores.
    David> Quitting and restarting Gnus fixed this.

    Brad> The BBB has a three hour inactivity timer on your login
    Brad> token.  If you haven't sent any ratings or received any
    Brad> predictions for more than 3 hours, you should see a message
    Brad> telling you to log-in.  Just M-x bbb-login will do the
    Brad> trick.

OK.  The symptoms I were seeing involved a message that mentioned
timeout that was very quickly overwritten so I wasn't sure what had
happened.  I'll check this out next time it times out.

Interestingly, the first day I started using GroupLens, most of the
articles in the GroupLens groups had predictions.  Today, none of the
new articles seem to have predictions (I can still see the predictions
for old articles).

Presumably, with GroupLens you should be able to read only articles
with high predictions (plus any you might select for other reasons).
You will then only be rating a relatively small number of articles.
Do you expect that with a large user community, each user will only
need to rate a very small percentage of articles but there will still
be enough ratings for the system to work well?

Again, thanks for GroupLens.

David

--
David Allen (davida@lsl.co.uk)		   Laser-Scan Ltd, 
Principal Software Engineer		   Science Park, Milton Road
Tel: 01223 420414   Fax: 01223 420044      Cambridge, CB4 4FY, UK