On Sun, Aug 31, 2003 at 01:47:01AM -0400, Walter Landry wrote: > Steve Langasek <vorlon@netexpress.net> wrote: > > On Sun, Aug 31, 2003 at 12:26:04AM -0400, Walter Landry wrote: > > > Based on faulty information, the Release Manager told them not to > > > bother. Now they should bother. > > Where was this said? The only statement I've seen is that these > > bugs will not be considered blockers for sarge. Do you mean to say > > that the maintainers of all the affected packages only fix the RC > > bugs against their packages, and ignore everything else? > That is why they are called release-critical. They have to be fixed > before the release. Other bugs do not. The release manager has some > discretion to decide that a bug isn't _really_ release critical, but I > didn't think he could just ignore the Social Contract. > That is my big question, which no one seems to want to answer. Is it > ok for the Release Manager to ignore the Social Contract? These > documents are not going to become free in the forseeable future. Is it ok for the maintainers of the packages that contain GFDL documentation to ignore the Social Contract? They have also agreed to uphold it; it shouldn't require a mandate from the release manager to get these bugs fixed. (Even treating the bug as RC does not guarantee the Social Contract has been upheld, as it only guarantees the bug will not exist in the release -- possibly by removing the package from testing and leaving it, bugs and all, in unstable.) > > The severity of these bugs has not been changed; they are still > > considered serious bugs, and they still need to be fixed. > And yet they are somehow not really serious bugs, since serious bugs > are, by definition, release-critical. No. Serious bugs are, by definition, violations of "must" requirements in Debian Policy. This makes them release-critical by *default*, but this is not the first time that a serious bug has been ignored for a release. > > Nothing stops the maintainers from working on them between now and > > the freeze date if they have the time for it. Nothing stops you > > from working on them, if you feel this is important to resolve prior > > to release. But if no one is willing to work on them, your claim > > that there won't be a significant delay seems rather ephemeral. > The problem is not manpower, it is willpower. gcc, for example, has > already done most of the work to fix this bug. All that really needs > to be done is for someone with authority to tell them to apply it. Does this mean that the gcc maintainers don't agree with this list's interpretation of the GFDL, or that they don't regard this as a high priority between now and the release? Does the patch have negative side effects that leave the maintainers reluctant to apply it (such as leaving sarge without any gcc manual at all, even in non-free)? -- Steve Langasek postmodern programmer
Attachment:
pgpnZpRJOGLc8.pgp
Description: PGP signature