Re: On Bugs, take 2
> * We need some way of marking out which problems make packages
> unsuitable for release (ie, we're better of not releasing the
> package than releasing it with that problem).
>
> * We need some way to differentiate other problems to work out
> which are more important to focus efforts on fixing those.
>
> At the moment the "important" severity is often being used for the
> latter although it's theoretically intended for the former, which makes
> it somewhat useless.
I agree.
> So, how about splitting them? What would people think of a bug tag:
Again, this makes sense; people's habits do not easily change.
> brownbag -- A bug that makes you want to put a brown paper bag over
> your head. Something that would be annoying to release
> with, but not necessarily unacceptable. The package
> maintainer has the final word on whether a bug is a
> brownbag bug or not. Brownbag bugs will not affect
> whether a package is eventually released or not.
>
> ?
>
> Would this make clarifying "important" to be:
>
> important -- A severe violation of Debian policy (that is, a violation
> of a "must" or "required" directive).
> more acceptable to people? (I say "clarifying" because that's what I was
> using to differentiate "important" and "normal" bugs during the potato
> freeze) Again, please note that filing an "important" bug is *not*
> something people should be doing naively: it's not a voting mechanism,
> it's not for opinions, it's for saying that this package must *not*
> be released in its current state, under almost any circumstances.
This does not, IMHO, fix the problem. Those people who are naively
reporting bugs (and who this change is intended for) are not likely to
read this message or any documentation on this: ``brownbag, what's that?
Hmm, must be important.'' Thus, important should be left in its current
``popular'' meaning, i.e. what you are calling brownbag, and a new name
should be given to important bugs, e.g. violation.
-Neal
--
Neal H Walfield
University of Massachusetts at Lowell
neal@walfield.org or neal@cs.uml.edu
Reply to: