i-scream documentation viewer
minutes-20010305.txt
Minutes of meeting 05/03/2001 @ 2pm
Location: UKC Computer Science Meeting Room
Present: ab11, ajm4, pjm2, tdb1
This meeting started with the long awaited demonstration to
John... except he brought a few colleagues along :) With our
captive audience, we demonstrated the past 20 odd week's
work, and fortunately got lots of positive feedback. This
certainly boosted the confidence of the group as it made all
the effort seem worthwhile. One comment worth nothing was: -
"I wasn't expecting it to be quite so useful!"
After the demonstration, the group proceeded to discuss
where to head next. Particular issues on the agenda were the
webpage front end, and the alerting mechanisms.
Both AJ and Tim had spent a couple of very late nights
writing the alerting mechanisms (the local clients) for the
server. They noted that the logic behind it all was a lot
more complex than had been anticipated, and that it really
was a major part of the system. This work had been mostly
done for some basic data (CPU levels) and just needed to be
expanded to cover the rest. This was to be done over the
following week.
Ash was pleased to report that the host was now working,
although there still existed a problem with packet size. All
the other issues with it crashing etc had been tidied up. He
again noted that he was keen to get some work done on the
webpage frontend.
Paul said he had completed most tasks, including the winhost
and the DBReports. He volunteered to do some work on the
webpage frontend.
The webpage frontend was something that needed doing. The
group agreeded it should be broken down into sections on
latest host data, latest alerts, and historical information
(db reports). These would all be tied together into one
easily navigatable interface, with a nice summary page. This
would tie closely with the alerting mechanisms discussed
above, so no plans for the actual implementation details
could be made at this stage.
Another issue that evolved from this was the host sampling
data and sending averages, and alerts. Ash said logic was in
place to start this, and he had it planned. AJ pointed out
that the initial design for the host was to be as
lightweight as possible, and this would just complicate it,
especially when the same work is carried out in the server.
The topic wasn't completely resolved, although the group
seemed to agree that it shouldn't be a priority with the
short time left.
The final issue was documentation. This would be looked at
over the week, and maybe another meeting arranged later on
to plan it.
There will be a meeting with Ian at 2.15pm on Wednesday, and
the usual meeting on Monday at 2.15pm. There may be an extra
meeting in the week to discuss documentation.
