Bug#231862: closed by John Ferlito <johnf at inodes.org> (libvorbis0a: oggenc still dies sometimes with floating point exception)

[Helge Kreutzmann]

Hello John,
On Mon, Apr 05, 2010 at 11:48:18AM +1000, John Ferlito wrote:
> On Sun, Apr 04, 2010 at 04:42:07PM +0200, Helge Kreutzmann wrote:
> > Hello John,
> > On Sat, Apr 03, 2010 at 03:33:08AM +0000, Debian Bug Tracking System wrote:
> > > This bug is ancient and is most likely fixed. Feel free to re-open if
> > > you don't think that is the case.
> > 
> > What kind of bug maintenance is this????
> 
> <snip>
> 
> I find it unfortunate that you find the need to simply jump to
> conclusions and send such an incendiary email in a public forum,
> rather than perhaps send me a quick private email to find out why I
> went down this path.

Sorry, I was probably a bit quick in my reply. But please realise that
I tried to supply a good bug report (find a reproducible test case,
supply a back trace, keep the sample for further debugging) and then
heard nothing (!) in this bug, not even a simple "ack, I'll look at it
later". 

> I have only recently taken over maintenance of these packages in an
> effort to try and clean up the mess that has been accumulating for
> many years. With a goal to try and keep them maintained much better,
> seeing as I do have some time to devote to this whereas past
> maintainers may not have.

Ok, then please accept my apologies. I was not aware of this and your
previous e-mail did not indicate that you are the new maintainer of
this package. It's great that you take care of this package and its
bugs.

> libvorbis, vorbis-tools and libao have not been maintained in an
> extremely long time. I have spent the past two days of my Easter
> creating new packages since upstream have finally after many years
> released new versions.

Great! Hope they make it for Squeeze.

> This required me to go through approx 60 bugs, most of them dating
> back to 2004 and 2005 which are simply not relevant to the current
> code base.

This probably is not a nice task. I remember that I once took over a
package and tried to hunt down reported bugs but the submitters had no
longer the hardware to reproduce. So the only course of action I had
was to tag them "moreinfo". Not nice if you try to go down to zero
bugs (to get the best for users). In your case probably worse, because
of the number of bugs (and to get an overview of "active" bugs and
"passive" ones).

> For most bugs including this one I took the time to have a quick look
> at the source code and change logs to try and ascertain whether or not
> the bug was likely fixed. This involved me taking one of a
> couple of steps

Good to hear, I wasn't aware of this.

> a) Pushing the bug upstream where I confirmed it was still an issue
> 
> b) Closing bugs that were unreproducible and no response had been
> heard from the original submitter when queried
> 
> c) Closing bugs where it seemed unlikely there was a problem any
> longer asking for them to be reopened if it was believed the problem
> still existed

I can perfectly understand your proceeding here, but I don't know if
this is the right course of action. You could tag it "moreinfo" and at
least send a query (and then push it to the b) category). I personally
don't know what the proper course of action with case b) actually is.
(And I don't have time atm to look at policy). Personally I would be
reluctant to close bugs which I don't know if they are really closed.

> d) probably some others I can't remember right now

> For this bug and many others, (Keep in mind this bug was originally filed
> back in 2004, that's 6 years ago now), after a quick look at the code
> and considering the amount of time I came to the conclusion that
> it was extremely unlikely that the bug still existed and chose option
> c).

Probably right, given my reasoning in the last e-mail, but maybe a
quick query would have been nice, e.g. sending a note to all packages
in category c) stating that you believed they were fixed and if they
could try to confirm this with your latest package. And if no response
come in in a reasonable time frame, you could move them to b).

> Thank you very much for taking the time to indeed confirm that this
> bug is most likely no longer relevant.

When I report bugs I try to follow through as much as possible,
because my intend is to get the issue solved.

So I perfectly understand your reasoning, but for next time (maybe
another lonely package :-)) the path could be improve as well. And
sorry if I reacted to strongly, and good luck and (hopefully) fun in
maintaining the vorbis family!

Greetings

           Helge
-- 
      Dr. Helge Kreutzmann                     debian at helgefjell.de
           Dipl.-Phys.                   http://www.helgefjell.de/debian.php
        64bit GNU powered                     gpg signed mail preferred
           Help keep free software "libre": http://www.ffii.de/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.alioth.debian.org/pipermail/pkg-xiph-maint/attachments/20100405/a2a291b3/attachment.pgp>