Am 03.05.2007 17:17 schrieb Julien Cristau: > On Thu, May 3, 2007 at 16:48:10 +0200, Bastian Venthur wrote: > >>> reportbug provides a way to control how bugs are reported on a >>> particular package, by adding files in /usr/share/bug/$package. However >>> reportbug-ng doesn't seem to use that information, which means that bugs >>> reported with it lack very important information. In particular, >>> without the output of /usr/share/bug/$package/script, many reports are >>> useless. >> I really doubt that. Just a few packages use this feature and if you as >> a maintainer are really missing some information, you can always ask the >> submitter for it. > > Which means: > - lost time for me > - lost time for the submitter > - more often than not, this information is never given > > If I'm requesting that information, it's not to annoy you, it's because > without it bug reports are useless, and it's easy for any bug reporting > tool to include it. If you find that a package adds useless information > in its script, then file a bug against that package, but the bug > reporting tool is not the place to make that call. > I don't consider this as a wishlist bug.
I do. You don't report a bug in rng but requesting a specific feature. >> You say that bugs with rng lack of important information and that many >> of them are useless. Could you please backup this claim by a few >> examples? > > Any bug reported against the xserver-xorg* packages. Any? Those are *all* xorg related bugs I've ever reported and in *none* of them the output of the script was relevant: #326755 #322543. I picked three random bugs from xserver-xorg using rng: the first one in the list: #211765 - some license problem, output was not relevant one from the middle: #354363 - no answer to the bugreport so far the last one: #369900 - looks like the output of the script was not relevant at all I don't doubt that this feature *is* actually useful, I just don't believe that it is relevant for *all* or even the majority of bugreports. I may implement this feature in the future (although it's pretty low on my priority list), but if I do, it would be fully optional to include this info or to read the presubj. Cheers, Bastian -- Bastian Venthur http://venthur.de Debian Developer venthur at debian org -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
