quoted with permission:

00:15 < madduck> are there any BTS people around?
00:15 < ari> he quit freenode
00:15 < ari> don kinda works on bts
00:15 < madduck> right. dondelelcaro?
00:15 < ari> yup
00:15 < madduck> ari: (that was supposed to trigger his client... ;^>)
00:15 < ari> i figure
00:15 < dondelelcaro> madduck: ?
00:16 < madduck> wow, it works. :)
00:16 < madduck> dondelelcaro: just had a quick thought and wanted to get
                 aj's/yours/someone's opinion.
00:16 < madduck> dondelelcaro: submitters aren't (yet) subscribed to their bugs
                 by default.
00:16 < dondelelcaro> yes
00:16 < ari> well there's a reason -submitter is separate
00:17 < madduck> how difficult would it be to add an X-* header to allow a
                 submitter to be automatically subscribed?
00:17 < madduck> like X-Debbugs-Cc I mean.
00:17 < madduck> X-Debbugs-Subscribe: please
00:18 < dondelelcaro> madduck: we need a shared secret between the bts and what's
                      doing the list subscription; I can do it right now if
                      people don't mind dealing with the subscription confirmation
00:18 < madduck> i think the confirmation is good and necessary.
00:19 < madduck> is it not too much trouble to implement? i suppose it's a
                 headercheck and an email sent, huh?
00:19 < dondelelcaro> no, it's fairly simple.
00:19 < madduck> even simpler?
00:19 < dondelelcaro> just a few lines in process.in would do it
00:20 < madduck> this may just be a turnaround record for IRC communication in
                 Debian wishlist procedure.
00:20 < madduck> :)
00:20 < dondelelcaro> well, that would require me to actually implement it and
                      test it... which will take a while;
00:20 < madduck> yeah hey... no stress.
00:20 < madduck> at least i got the idea out?
00:20 < madduck> want me to file a bug against b.d.o or debbugs?
00:20 < dondelelcaro> debbugs is probably best
00:21 < madduck> consider it done; can i just quote?
00:21 < dondelelcaro> but yeah, file the bug I'll stick it on the todo list...
                      I'll probably nail it with the rest of the "do more
                      intelligent stuff with pseudo headers" things
00:21 < dondelelcaro> madduck: sure, go right ahead

Any news on this? Imagining how much time has been wasted because
submitters didn't see followups to their bugs makes my mind boggle...

I'd love to know whether there is an intention of fixing this.

Thanks!

Kumar

As this bug gets some attention and some people refer to it
as "auto-subscribe" by default, I just want to add my vote that
auto-subscription should be an option and never a default.

It's one of the biggest advantages of the bts that it does not
pester people doing us the service of reporting bugs, but keeps
a clear separation between things related to the bug and things
to concern the submitter with.

Some way for bug-submitters to request subscription and also see
internal discussion is nice, but that does not change the duty of
the maintainer to CC the submitter (or some more inteligent
address only reaching the submitter once when he is subscribed)
whenever sending a message that needs something from the submitter.

Hochachtungsvoll,
	Bernhard R. Link

Today again, I unfreezed a bug which was freezed because the submitter
was unaware he was asked a question (bug #493040, and this one was RC!).
Indeed, people often forget their "duty" of CC-ing the submitter, that's
why I strongly support this pragmatic idea.

A solution may be to use a different sumission address if the submitter
wants to be involved in the process of fixing his bug,
submit-cc@bugs.debian.org, for example, and interactive bug reporting
tools should suggest being kept aware of the discussion on the bug.

It should not be too much work, would not bother users, and would
greatly enhance the efficiency of the bug resolution process.

Regards,
Nicolas

Not being auto subscribed to bugs you submit is truly one of the most
bizarre and frustrating aspects the debian bts. As far as I know, the
debian bts is only bug tracking system that works like this.

Avery

-- 
Alexander Galanin

Hi,

Any progress on this 3.5 years-old bug?

Quoting from comment #10:
"Any news on this? Imagining how much time has been wasted because
submitters didn't see followups to their bugs makes my mind boggle..."

Why spending time to implement other features on b.d.o., while this bug
still makes MUCH MORE time being lost?

Analogy: suppose we need to optimise the execution time of a program.
All the programmers run a perf program to see what functions of the
program take time, and fix the top functions.  All the programmers will
prefer to fix a top function instead of 50 functions at the bottom of
the list.

(I am not trying to blame anybody!!!, just to raise the importance of
this bug.)

Cheers,

You don't get to determine where I spend my time. I've outlined what
needs to be done in a mail to -devel, and the work is in progress.
Patches accepted; you can see the work on this branch in
http://bzr.donarmstrong.com/debbugs/.

[Needless to say, I won't be responding to further mails about this,
as it will only delay the implementation of this feature.]


Don Armstrong

In case others get here while trying to figure out how to automatically
subscribe to bugs that you submit, I'll post some links to get you on your
way a little faster.

Here is a conversation on the debian-devel list:
http://lists.debian.org/debian-devel/2009/09/msg00422.html

And the end result is, various people have strong opinions on the correct
functionality, and so there isn't a way to make everyone happy.

My solution is using the auto-subscribe script in the devscripts package,
which solves it for me.

/usr/share/doc/devscripts/examples/bts_autosubscription.procmail

I modified it slightly, since I don't use the control@ address that much.
I probably should have some other check that looks for the hostname or
something, to avoid people being able to get me to send mails to
[0-9]-subscribe@bugs.debian.org, but I suppose that isn't very interesting
for a spammer, other than in a DDOSing kind of way.


# This handles initial submissions ("Thank you for the problem
# report...") and also followups ("Thank you for the additional
# information...").
:0
* ^To: me@my\.email\.address$
* ^From: owner@bugs\.debian\.org \(Debian Bug Tracking System\)$
* ^Subject: Bug#\/[0-9]*
|echo |mail "$MATCH-subscribe@bugs.debian.org"

:0
* ^From: [0-9]+-subhelp@bugs\.debian\.org$
* ^Subject: Please confirm subscription to [0-9]+@bugs.debian.org
* ^Reply-To: \/.*@bugs\.debian\.org$
|echo |mail "$MATCH"

:0:
* ^From: [0-9]+-subhelp@bugs\.debian\.org$
* ^Subject: Subscription to [0-9]+@bugs.debian.org successful$
Mail/bug_subscription_success

Dear Customer,

UPS courier was unable to contact you for your parcel delivery.

Postal label is enclosed to this e-mail. Please check the attachment!

Respectfully,
Jessie Rowe,
UPS Senior Station Manager.

If most people are using reportbug to file tickets, reportbug has a
novice mode that can probably be used to decide whether to insert a
`Control: subscribe -1 !` to the submit message.

So should this be reassigned to reportbug?

regards
Afif

على ٢٨‏/٥‏/١٤٤٠ هـ ‫١:٥٧ ص، كتب Afif Elghraoui:

silly me--- I mistook this for a control command since I usually use
devscripts' `bts subscribe ...`.

Never mind my previous message and apologies for the noise.

regards
Afif

YA zainteresovan v finansirovanii vashego investitsionnogo proyekta za schet kredita ili partnerstva.