That need to close developers?
Vincenzo Ciancia
ciancia at di.unipi.it
Thu Sep 20 13:39:23 UTC 2007
On 20/09/2007 Colin Watson wrote:
> As a developer, I wish I didn't have to spend time checking my
> bug mail just to make sure that well-intentioned but mistaken triagers
> aren't taking items off my to-do list that I want to stay there.
>
<sarcasm>
As we are in topic, I suggest that, if a complete bug report stays
opened for more than a month without any reply, and a developer is
reading it, then the developer should be automatically closed and marked
as invalid.
</sarcasm>
Seriously speaking, in my opinion there are both many bug reports that
need info and nobody will provide that, and many bug reports that are
complete or need further assistance by a developer to be completed, but
survive for months, years and releases without receiving attention.
While I can perfectly understand that one marks things as "to do" and
leaves'em there until he/she has time, on the other hand users should
have means to get noticed when a problem is surviving for too much time
without being taken in account.
Suppose one wished to automatically mark as invalid bug reports when
obviously no one will take care of providing needed information. Then
one could arrange things in launchpad as follows:
1. the bug is marked as "needsinfo" or something like that
2. after one month, if no comment has been posted on the report, then
it's closed
3. any comment on the report removes the "needinfo" tag during the month
This can be dually extended to developers and open requests:
1. users can mark bugs in various ways (see below)
2. after one month, the bug is marked as high priority
3. any comment on the report removes the mark during the month and
eventually the priority is reset
Marks on bugs can be e.g. "needshelp", meaning that user needs help to
provide a backtrace or more information (not all programs are equally
easy to debug, for example I never succeded in providing an openoffice
backtrace), or "goodinfo" to signal that reporters have done their best
to provide any possible information, or also "containsfix" to signal
that a fix has been suggested or provided in a comment.
I don't expect users to abuse of such a mechanism, just as I don't
expect developers to close bugs on purpose just because they don't have
time to deal with them. And if it does not work, the feature can always
be removed.
Vincenzo
More information about the Ubuntu-devel-discuss
mailing list