Re: ANNC: New Debbugs
On Tue, Jan 21, 2003 at 02:13:00AM -0600, Adam Heath wrote:
> On Tue, 21 Jan 2003, Ola Lundqvist wrote:
>
> > > Hmm. Interested idea. But perhaps those people should use reportbug or bug.
> >
> > Yes maybe. But sometime (even I had problems using bug and reportbug
> > before I learnt the EMAIL env variable) such tools can be hard to use,
> > at least for a novice. I have not looked at them for half a year or
> > more but "back then" they was a bit hard.
>
> Then file bugs on them to have them improved.
Well, yes. :)
Actually, one of the reasons why I want this feature is becuase I like
the debian bug system so much that I want to use it in other projects.
> > Ok, but at least the initial reporter must be automatically subscribed to the bug
> > in order to get all the information to the bts. If not, the reply-to header
> > will not be set properly.
>
> Why would the initial reporter always be concerned about everything that
> occurs for a bug? And what does that have to do with setting reply-to?
I want to solve (or at least help as much as possible) the problem withsubmitters,
maintainers, etc that do not send their additional information to the
bts.
> Convince me properly.
I'll try.
Ok, let me give a usecase. And some improvements to my proposal.
Maybe I can lurk out my own specifications on how I was about to
write my own such system.
The only problem I can see is spam. Right now the pseudo header
is one thing that prevents spam-mailers to introduce new bugs.
The From:-line should not be changed so that mail can be
sent to the origin directly without sending information to the
bts, but that should not be the default.
1) User submits a bug. He/she is a newbie and sends an email
to foopackage@bugs.debian.org
1.0) Bug number created.
1.1) The user will be subscribed to bugnr as type submitter.
1.2) The package is registered as a subscriber to this bug as
type package (this might not be necessary as it is implied).
1.3) An accnowledge is sent to the submitter and informs that
he/she is subscribed to this bug and informs how to unsubscribe
from further information. This mail contains Reply-To: bugnr-maint@
header if he/she want to add more information about the
subject to the maintaier of the package.
2) The maintainer gets the bug-report with Reply-To: bugnr@
set. One alternative is to set bugnr-submitter@ instead
so the mail is sent to the submitter. I do not know if this
is an ok way.
3) The maintainer (or co-maintainers) reply to this address
3.1) Mail is sent to all (or just submitter if -submitter is
specified).
4) Submitter recieves an email with reply-to set and then
replies.
I'm not sure ff bugnr@ or the alternative (bugnr-subimmer and
bugnr-maintainer) should be used, becuase I have not tested
if there are other use-cases.
I think this whould be a neat system for debbugs and solves
the problem with users (especially users) that reply just to the
maintainer and not to the bug-number.
Is this a better description of the system.
Regards,
// Ola
--
--------------------- Ola Lundqvist ---------------------------
/ opal@debian.org Annebergsslingan 37 \
| opal@lysator.liu.se 654 65 KARLSTAD |
| +46 (0)54-10 14 30 +46 (0)70-332 1551 |
| http://www.opal.dhs.org UIN/icq: 4912500 |
\ gpg/f.p.: 7090 A92B 18FE 7994 0C36 4FE4 18A1 B1CF 0FE5 3DD9 /
---------------------------------------------------------------
Reply to: