> < ^ Date: Thu, 29 Apr 1993 17:20:24 +0200
> < ^ From: Joachim Neubueser <joachim.neubueser@math.rwth-aachen.de >
^ Subject: bugs and trouble

Dear GAP-forum,

After some discussion, we want to introduce a new address

gap-trouble@math.rwth-aachen.de

The idea is the following. We have observed the discussions and
reports in the GAP-forum as well as in the mail that some of us have
got privately about GAP. These fall roughly into four categories:

1. Reports on what I like to call "real bugs", i.e. on instances where
GAP produces a wrong answer to a problem, an answer that the user
might even take for granted and thus might be led to wrong
conclusions. The wrong abelian invariants, on which Daniel Ruberman
reported recently, were such a case. (in which Martin fortunately
could at least clarify that this was a rather exceptional
malfunction).

*We urge that each such case should by all means and as soon as
*possible be reported in the GAP-forum in order to minimize the danger
*that somebody is led to wrong conclusions.

2. Reports on annoying behaviour of GAP such as that of GAP after
interrupting while printing that Andries Brouwer reported this
morning. Such things are clearly annoying, they may cost the user's
time, but at least (s)he cannot be led to wrong mathematical
conclusions. The same applies to inefficiencies. We have no objection
at all if these are sent to the GAP-forum, however more such reports
seem to be sent directly to a private address in Aachen, most often to
Martin Schoenert. We ask that such reports if not sent to the
GAP-forum are rather sent to the above address "gap-trouble". Letters
sent to that address will automatically be distributed to a small
group of people who will jointly try to deal with the problems. This
group will be started with Martin Schoenert (of course), Frank Celler,
Thomas Breuer and myself, but whoever is willing to help us with the
task to settle such deficiencies is invited to join the group (please
just tell us to be put on the address list). Letters sent to
"GAP-trouble" will get an individual reply to the sender and in
addition we intend to give a report from time to time to the GAP-forum
summarizing the troubles reported and our plans to deal with them,
which may range from "yes will be fixed in the next patch" over "yes
but has to wait for the next version" to "yes, but ...". We hope that
by this scheme we cannot only distribute the workload a little bit
more evenly, but also to minimize times of no reply when some of us
are absent from Aachen.

3. Installation problems. The flood of them occurring in the GAP-forum
after the release of GAP 3.2 even led to the proposal of a second
forum. While we did not follow this suggestion, we think that if in
the future such problems are also sent to "gap-trouble" and we report
only on important and more often asked questions in the GAP-forum, we
achieve some better organisation in this respect, too.

4. Questions "How can I do in GAP ...? " or "Can I do..?" These we
would like to see further in the GAP-forum and we hope that often, as
in the past, helpful answers will come from members of the forum
outside of Aachen.

The address "gap-trouble" will be operational from Friday, April 30.
We hope that it will be a useful innovation and ask you to use it.

Joachim Neubueser


> < [top]