Re: Re[2]: Separate list for newbies
> As for the volume of the list my solution is simple: I archive in
> gz files every single post to the list, and I have over a year's
> worth of such archives. There's nothing new under the sun and
> chances are I am not the first person to ask a question. This is
> something simple anyone could do, unless they are on a system where
> their disk space is severely curtailed, such as at a school,
> etc. That's why the list is archived at debian.org.
> The bottom line is: if a reasonable post to this goes completely
> ignored that's a good indication that no one else knows the answer
> either!
Maybe this is just completely impractical (I've just been through the
"dreaming up projects for next year's students" process), but how
about:
A 'bot that scans all "new" posts to debian-user and automatically
does a search of the list archive (plus FAQ etc), based on the
subject and a scan of the body text for keywords, and then sends a
response listing links to the top 5 or so results. It would have to
be bright enough to filter out its own responses from future
searches.
If these "automatic" answers are incorrect, out of date, etc,
somebody will quickly post better answers as followups, and next
time around, the system will favour these better answers.
This idea is based on little more than the vague impression that if I
were to systematically submit a suitably condensed version of each new
question to a list archive search engine, I would get pretty good and
useful set of responses.
Benefits:
For the genuine FAQ's the system would quickly home in on the
"correct" answer, and each such question would be quickly and
unobtrusively answered. Nobody else need bother responding, unless
they have new information rendering the standard response obsolete,
thus cutting down on repetitive noise.
For all questions, the first automatic response would be a quick
summary of previous threads on the subject.
Problems:
It might make the problem worse: after all it would be sending out
lots of extra messages to a list that is already busy.
The system might be no good at choosing suitable keywords for the
search, so the results are useless.
Nobody has the time to write such a system (I certainly don't) so it
never gets done.
--
Gilbert Laycock email: gtl1@mcs.le.ac.uk
Maths and Computer Science, http://www.mcs.le.ac.uk/~glaycock
Leicester University phone: (+44) 116 252 3902
Reply to: