bug#10134: acknowledged by developer (close 10134) | [debbugs-tracker] Processed: close 10134

bug#10134: acknowledged by developer (close 10134) | [debbugs-tracker] Processed: close 10134

[Drew Adams]

Another bug closure with no explanation mail sent to the filer.  Just a reminder
that this is not really kosher.

Citing Bob Proulx, from the last time you did this: 

"See the documentation for the 'close' command to see that
 it explicitly calls out this use case as a problem and asks
 maintainers to ensure that they communicate, probably by
 sending a separate message, why the bug is being closed.
 See the section 'close bugnumber'
 http://debbugs.gnu.org/server-control.html."

(Yes, I read the bug thread and I realize the bug has now been fixed.)

FYI, these are the only closure messages I received, both automatic:

> -----Original Message-----
> From: GNU bug Tracking System
> Sent: Saturday, February 11, 2012 6:06 AM
> To: Chong Yidong Cc: tracker@debbugs.gnu.org
> Subject: [debbugs-tracker] Processed: close 10134
> 
> Processing commands for control@debbugs.gnu.org:
> 
> > close 10134
> bug#10134: 24.0.91; Typo, (elisp) `Text from Minibuffer'
> bug closed, send any further explanations to
> 10134@debbugs.gnu.org and "Drew Adams" <drew.adams@oracle.com>
> 
> > thanks
> Stopping processing here.
> 
> Please contact help-debbugs@gnu.org if you need assistance.
> 
> GNU bugs database, http://debbugs.gnu.org/



> -----Original Message-----
> From: GNU bug Tracking System Sent: Saturday, February 11, 2012 6:06 AM
> Subject: bug#10134 acknowledged by developer (close 10134)
> 
> This is an automatic notification regarding your bug report
> #10134: 24.0.91; Typo, (elisp) `Text from Minibuffer',
> which was filed against the emacs package.
> 
> Thank you for your report, which has now been closed.
> You can view the full report at
> http://debbugs.gnu.org/cgi/bugreport.cgi?bug=10134
> 
> If you require further information, please followup to 
> 10134@debbugs.gnu.org.
> 
> debbugs.gnu.org maintainers
> (administrator, GNU bugs database)

bug#10134: acknowledged by developer (close 10134) | [debbugs-tracker] Processed: close 10134

[Chong Yidong]

"Drew Adams" <drew.adams@oracle.com> writes:

> Another bug closure with no explanation mail sent to the filer.
> Just a reminder that this is not really kosher.

I hereby declare, under the religious authority vested in me by the
Church of Emacs, that closing bugs with no explanation mail does not
violate kosher laws.

> (Yes, I read the bug thread and I realize the bug has now been fixed.)

... especially when the objection is this silly.

bug#10134: acknowledged by developer (close 10134) | [debbugs-tracker] Processed: close 10134

[Drew Adams]

> > Another bug closure with no explanation mail sent to the filer.
> > Just a reminder that this is not really kosher.
> 
> I hereby declare, under the religious authority vested in me by the
> Church of Emacs, that closing bugs with no explanation mail does not
> violate kosher laws.

http://debbugs.gnu.org/server-control.html:

 close bugnumber [ fixed-version ] 
 Close bug report #bugnumber. 
 A notification is sent to the user who reported the bug, but
 (in contrast to mailing bugnumber-done) the text of the mail
 which caused the bug to be closed is not included in that
 notification. The maintainer who closes a report should ensure,
 probably by sending a separate message, that the user who
 reported the bug knows why it is being closed.

Not providing an explanation for the closure requires the bug filer to look up
the bug number, consult the bug thread on line, and reread the thread looking
for an explanation.

When you close a bug it might be obvious to you why it was closed, but it is not
obvious to a user who receives only an automatic control message saying it was
closed.  The only information available to the user in that message is the bug's
number and subject line.

> > (Yes, I read the bug thread and I realize the bug has now 
> > been fixed.)
> 
> ... especially when the objection is this silly.

Silly and all a big joke to you perhaps.  But to obtain that knowledge of what
the closure meant (the bug was fixed) I still had to look up the bug thread and
reread it.  If the automatic message had said that it was fixed then I would not
have needed to do that.  Perhaps that part of the automatic messaging could be
improved.

If a bug has been fixed, knowledge of that fact is often all the filer needs.
If it has not been fixed (e.g. was declared not-a-bug or wishlist or wont-fix),
then additional explanation is typically needed.

bug#10134: acknowledged by developer (close 10134) | [debbugs-tracker] Processed: close 10134

[Jason Rumney]

"Drew Adams" <drew.adams@oracle.com> writes:

>  The maintainer who closes a report should ensure, probably by sending
>  a separate message, that the user who reported the bug knows why it
>  is being closed.

Which is exactly what happened, according to your comment:

> (Yes, I read the bug thread and I realize the bug has now been fixed.)

bug#10134: acknowledged by developer (close 10134) | [debbugs-tracker] Processed: close 10134

[Drew Adams]

> >  The maintainer who closes a report should ensure, probably 
> > by sending a separate message, that the user who reported
> > the bug knows why it is being closed.
> 
> Which is exactly what happened, according to your comment:
> 
> > (Yes, I read the bug thread and I realize the bug has now 
> > been fixed.)

Uh, nope.

I wrote that in my message that pointed out that I received no message
explaining the closure.  I wrote it _after_ getting the automatic closure
message, looking up the bug number, and consulting the thread.  After all that,
yes, I realized that bug #10134 was one that had been fixed.

The point of the guideline is for the closer to ensure that the filer knows why
it is being closed before closing.  The point of sending a separate message is
to let the filer connect the reason/status with the bug number, without having
to traipse through the thread again.

The last message in the thread, before the automatic closure message, was
2011/12/25, over a month and a half ago.  Had the no-reason automatic message
come on that day or, say, Dec 26th, things might have been a bit different in
terms of my awareness of this bug and its status.

Between just 2012/01/01 and today there have been 280 messages for Emacs bugs I
filed.  Perhaps you can understand that a message letting me know that this bug
was fixed would have been helpful.

And as I said, in the case where a bug is fixed, that info alone is sometimes
sufficient for the filer, so perhaps the automatic closure message could include
that info.