On Wed, Aug 27, 2003 at 02:05:15PM -0500, Manoj Srivastava wrote: > On Wed, 27 Aug 2003 11:46:41 -0500, Steve Langasek <vorlon@netexpress.net> said: > > This is the sticking point, I think. Are we talking about resolving > > the possible problems *from* NMUing, or are we talking about > > resolving any problems that happen to show up after the NMU? I > How can one distinguish between the two, without > investigation? Often bugs are caused by the darndest things, and some > of the worst are the "can't happen" category bugs. Oh, surely you can't; but once a determination is made, I don't think the NMUer bears any more responsibility than any other developer to fix those bugs that were not actually caused by the NMU. > > absolutely agree that an NMUer is responsible for fixing any > > problems caused by the NMU, but I don't agree that NMUers should be > > held responsible for pre-existing bugs in the package -- whether or > > not they happened to be exposed by the NMU in question. > If you upload caused things to get worse for the users of the > package, you are responsible. If the upload has changed things for > the worse, you should try and fix it. The very least you *must* do is > monitor the package to ensure that your NMU is not causing problems. I agree with you but some of the assertions made in this discussion appear to go farther. > > If that's the case, I have no inclination whatsoever to NMU buggy > > packages -- I'd much rather file for their removal from the archive. > No one is holding a gun to your head. You are a volunteer, and > can't be forced to NMU. Quite. But I believe that doing NMUs *does* improve the overall quality of Debian, and I believe that putting NMUers on the spot over bugs they didn't cause would be a sufficient deterrent that Debian would suffer for it. I have no qualms about requesting package removals from the archive, but when I'm doing NMUs, I usually start with packages I *don't* think should be removed. > > Holding NMUers accountable for the quality of their uploads: yes. > Quite. If your upload caused the situation to deteriorate, > whether you deliberately caused the change that made it so or it was > inadvertent, you are responsible. I can't tell if we're agreeing here or not. :) Yes, if you make a change in the NMU that causes the situation to deteriorate, you are responsible; but a pre-existing FTBFS error in the package that just hasn't been filed yet isn't the result of a change made by the NMUer, and therefore the NMUer should not be held responsible. -- Steve Langasek postmodern programmer
Attachment:
pgpcWU4SZeBYx.pgp
Description: PGP signature