BerliOS bug tracker

Axel akoh... at gmail.com
Wed May 13 06:04:58 CEST 2009


> It would be enough for you to check messages.. check if they are serious
> bugs.. reproduce them.. keep the tracker clean and in order.. provide a
> pre-digested information to developers.. This would be already extremely
> helpful and I don't think that you need more than a user competency for
> doing this.

i would definitely second that, and as far as i can tell, this is
_exactly_ what would
be needed to make bug reporting work well to begin with. from my
personal
experience, i can add that this is much more interesting and
educational an
than what one may initially expect and it definitely does not need
much more
knowledge than some common sense and basic user knowledge. all that
needs
to be done is to follow up bug reports, file them to the bug tracker
if not there already,
collect input to reproduce the problem, and make it digestible to
somebody that
would know how to fix it. you will quickly find that people will
quickly fix problems
that are easy to check and at the same time you will learn a lot about
how
cp2k work, even without being a developer yourself, and it would
actually help
your research much more than you might expect. many people that i now
know
are contributing to open source software have started like this. who
knows, at some
point (and i'd expect it would be rather sooner than later) you may
find yourself in
the position to fix the bug yourself. this is what happened to me when
working
on CPMD. many times users would report a problem with a keyword. while
their
original report would often be useless, i would look up the keyword in
the manual
or the source, construct a simple test case, see whether i could
reproduce the
problem and look up the origin. many times the problem was easily
identifiable.
occasionally, especially at the beginning, the solution was not
perfect, sometimes
even wrong, but most of the time with the help of some more
experienced people
a solution would be found. quite often even, the eye of an outside
would see
problems, experienced people would not notice.

it is an unfortunate development of research that people nowadays are
frequently
looking for "the easy way out". the "just do something and forget
about it" solution
and not something that is sustainable. so you'll have to accept that
fact that if you
are willing to do the easy part, but don't want to get involved that
you'll be confronted
with what may  look like sarcasm to you. projects like cp2k survive
because of people
who _don't_ get discouraged by this. they rather get _motivated_. yes,
it is not easy;
yes, it is not always convenient; yes, it can be challenging at times;
yes, it can be
frustrating. if you look into the early posts in this group, you can
find me complaining
quite provocatively. however, i have been "converted". i may not
always agree with
choices that are made, or with how people handle issues, but for as
long as i
cannot do any better, i found it better to help wherever possible
rather than go
back into my corner and sulk.

from the point of somebody that has contributed a _lot_ of code and
effort to various
open source software packages, it is also quite frustrating to see
that people are gung-ho
about helping until it actually starts to look like some real effort.
at that point one sees
them claim that they are "not qualified" and "not developers" and
generally didn't mean to
offend somebody. i would appreciate it personally, if you spend some
time thinking about
how much effort of other people you benefit from and whether that
would actually
deserve a little bit more than just making some suggestions without
becoming active.

thanks in advance,
    axel.

p.s.: sorry for not continuing this privately, because i believe this
is a public issue.

> Regards,
> Teo
>



More information about the CP2K-user mailing list