Hi, Eddy Petrișor a écrit : > Hello all, > > After a discussion I had with gentildemon about bug > https://gna.org/bugs/index.php?11180 > I though it would be a good idea to get a feeling about what meaning people > consider that the > severity should have? > > > > Here follows my interpretation of them and what *I* know to be the general > practice. > > As I explained in the comments of the bug, in all the projects I have > participated (free software or > commercial) people associated the "severity" field mostly to the > "release-ability" of the code with > regard to that specific issue. >
I never participate in a project where "severity" field is associated with the "release-ability", because you also have bug on production software. In the projects I have participated, the "severity" field is used by the client and the project manager to agree on a level of support, on payoff, etc. For instance, a wish must be paid by the client, a normal bug enters in standard support, a blocker bug (which is generally marked with high priority!) could lead to refund or it will at least make the selling of next "tasks" more difficult for the project manager as reporting is done regularly. Following are *my* definitions :-) > Thus, for me the meanings of the severity levels are as follows: > - security - this is a security issue; fixing this bug itself warrants a > release; thus, when fixing > this type of bug is generally a good idea to make a release immediately (for > the latest releases, be > them stable or betas) > I agree with your definition of "security". > - blocker - this bug is that bad that we can't release the next version > without fixing this bug; in > other words, this bug blocks the next release > A blocker bug is for me a bug that leads to a crash or an infinite loop. Thus the user can't continue to use the software, even in degraded mode. He needs to restart the software (if possible). No release should happened with such a bug (or maybe a beta release *if and only if* it appears in very corner case!) > - important - this is an important issue and is preferable to fix this as > soon as possible; this bug > can be on the list of release goals of the next release > I mostly agree. So "important" means that the bug is not blocking, nor "security" bug. A new release is not needed when such a bug is discovered on the "production/stable" release. However, it should be probably fixed for the next release. > - normal - this is a bug that is a functional problem which can be annoying; > it would be a good > thing to fix this for the next release, but is not a problem if is postponed > - minor - a minor issue like spelling issues, small cosmetical issues; is > probably trivial to fix > such an issue and it would be nice to fix this for the next release > Well, your definitions are more precise that mines, that's good ;-) However, I'm still not sure that a "minor" bug is always trivial to fix... > - wish - usually a feature request; it might involve a lot of work, or even a > small amount of work; > definitely this is not an issue to stop the release. > I agree (and I complain why we can't move from bug to task without making hand-copy!) > I think this is the correct interpretation since all of severities relate > directly to the next release. > Here is the point :) For me, severities does not relate directly to the next release. Your view point is mostly a developer point of view. I have more an user point of view. For instance, in my mind, "minor" refers more to small bug that the user sees, but a small bug for a user may need huge amount of work for the developer. > Do people feel this is not the correct interpretation of the severities? I > got this feeling from > gentidemon since he associated a blocker bug with the fact the game itself is > blocking and doesn't > have to do at all with the release-ability of the code. > > > > What do people think? Which is the interpretation you think is more > appropiate/correct/fit for this > project? How do others interpret the severity field of the bugs? > > Could you please expand the explanation, if you map those severities > differently ? > yes ;-) Matt (gentildemon) _______________________________________________ Wormux-dev mailing list Wormux-dev@gna.org https://mail.gna.org/listinfo/wormux-dev
