[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: On maintainers not responding to bugs



Le Mon, Feb 26, 2007 at 04:37:10PM -0800, Don Armstrong a écrit :
> Allow me to quote from the OP:
> 
>    What do people look on the following idea: not allow packages to
>    migrate from sid to testing if they have unanswered bug reports
>    with severity >= normal?
> 
> Thus, if you have a package with any unanswered important or normal
> bugs, it will not progress. In order to assure propogation, you must
> respond rapidly to any bug that is filed with these severities, even
> though this has nothing to do with fixing the bug. This is a technical
> impedance to a maintainer getting useful work done, like fixing RC
> bugs in testing, and as such is not something that I can condone.

Dear all,

of course, it would sound reasonable to avoid to build unlearnable
systems with complex rules, and I think that the main interest of the
proposal is to put unaswered bugs under question. However, in the case
there would be something to be implemented, how about adding "priority =
low" to the criteria of the packages which will not automatically
migrate to testing?

The main problem I see with unanswered bugs is that nobody knows what is
the message they convey, as each maintainer has his own view on this.
Again, there could be technical workarounds, such as adding a
X-DD-Message to the control field of package, and let the BTS display
them. Then maintainers could explain brifely to the users their
workstyle, to avoid misunderstanding and wrong expectations.

For the packages which get many bugs reported, there could be mailing
lists on which help can be found. For instance, if your package deals
with medical or pre-clinical research, do not hesitate to bounce the bug
report to debian-med@l.d.o.

I am lurking on the ubuntu-science mailing list, and it seems that
they manage to motivate their communauty to do helpful work with the
bugs. For instance, it seems quite easy to get an @ubuntu email
account when participating to the life of the distribution. Of course,
@debian is used for a different purpose, but one could think about
similar welcome gifts.

Have a nice day,

-- 
Charles Plessy
http://charles.plessy.org
Wako, Saitama, Japan



Reply to: