[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian bugs belong to the Debian BTS

On Thu, 23 Oct 2003 00:30, Marc Haber wrote:
> >I don't get your point: in which sense does it help to use a script in
> >this case?
> I find it quite acceptable to ask a bug submitter if the bug still
> applies after a new upstream version has been packaged.

If the bug is difficult or painful to reproduce (eg the kmail bug when running 
out of disk space) and there is a major new upstream version then this may be 

For a trivially obvious bug (such as word-wrapping a URL in a message in a way 
such that another user of the same program can't click on it) and a minor 
upstream release (3.1.3 to 3.1.4) it seems hardly worth the effort.  In this 
example Chris could have used the same editor window he used to send the 
message to write a test message demonstrating the bug.

> I find it quite unacceptable to ask the submitter to file the bug
> directly with upstream, not bothering the Debian maintainer any more,
> since this hides the problem with the Debian package.

I believe that a large part of the value that Debian developers add to the 
community is in managing bug reports and fixing upstream bugs in our 

A Debian user will often be hesitant to report bugs upstream.  Upstream 
developers sometimes brush off bug reports with "it's probably an issue with 
your distribution".  The Debian developer can filter out the bad bug reports, 
collect all the relevant information, and then send a detailed and correct 
bug report upstream which will be taken seriously.  As there is often a lot 
of dialog between Debian developers and upstream developers any bug reports 
from the Debian package maintainer will be taken seriously even if it does 
not appear good at a first glance.

Also from the upstream development side, there is a process of training people 
in producing bug reports that contain all the necessary information for 
fixing a bug.  If 1000 people each report 1 bug and each bug takes 4 messages 
because information needs to be dragged from the user then that is a 
significant drain on the resources of upstream developers.  If a single 
person or small group of people filter the bug reports then it can really 
save some time.

A final issue is that when a Debian developer forwards the bug reports there 
is no need for the person who found the bug to get along with the upstream 
author.  There are a few upstream authors that I don't get along with.  In 
one case an upstream author told me that he never wanted to communicate with 
"you or your kind" again (*).  However he writes some good software that I 
use, and being able to get a bug report sent would be handy, but I take him 
at his word and don't intend to send him an email about a bug report or any 
other matter.  Forcing me to send bug reports upstream would effectively 
force me to not submit bug reports for the programs written by that upstream 

(*)  For reference this does not refer to a KDE person.

http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/    Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page

Reply to: