Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.

Bugzilla 3.6: Consistency, Completeness, and Usability

1 view
Skip to first unread message

Max Kanat-Alexander

May 7, 2009, 6:44:56 PM5/7/09
Okay, so I have a somewhat-radical proposal to make for Bugzilla 3.6,
that I've talked to LpSolit and a few of the other reviewers about
already, and it's this:

No new features for Bugzilla 3.6. Instead, we focus on fixing the
features we already have.

This means that we fix all the major HCI (usability) issues in
Bugzilla. We finish any incomplete features that we have (so adding
enhancements that actually just complete existing features is fine), and
we make Bugzilla behave in a universally consistent fashion.

There is work for everybody in this scheme! Refactoring the backend for
consistency is something we can work on. Fixing the frontend is good.
Updating documentation to actually be in sync with Bugzilla--that's
another good thing. Completing features that you have worked on in the
past would be another good item. There's something for everybody,
believe me.

The highest-priority items would be all the usability issues discovered
by pryzak's students, as tracked at this bug:

Examples of "completing a feature" would be: adding defaults to custom
fields, allowing multi-select custom fields to show up in the bug list,
indicating the sort order of a bug list, allowing email_in to accept
attachments, the ability to disable certain field values, finishing the
functionality of the "see also" field so that it's more useful--the
sorts of things that were "always in the plan" and that everybody
actually wants, but that we somehow never got around to.

"Fixing" a feature would be things like: allowing the patch viewer to
use other VCSes than CVS, because nobody uses CVS anymore, making the
way extensions and template hooks work to be more consistent, etc.

I've tagged some bugs that I think would be good to work on as part of
this plan with [3.6 Focus] in the whiteboard. Here is a URL to these bugs:

Note that I'm more interested in fixing things that actually affect UI
users at this point more than I am interested in fixing or completing
APIs, though some of that could possibly be done too.

I came up with this plan because I am absolutely tired of hearing
people complain about Bugzilla (try a daily Twitter search for
"Bugzilla" and see what you get), I think that we're racking up
incomplete features faster than we can complete them, and we're adding
new incomplete UIs faster than we can fix them. We already have more
features than any other bug tracker in the world--now we should make
them *nice*.

Feel free to tag some bugs as [3.6 Focus] yourself. If you're not sure
that a bug would qualify, just ask me. Basically, any long-term "wow
this is an annoying and enormous problem" bug would be qualified.

Also, to be clear, I'm not going to forbid anybody to work on a whole
new feature. You're still welcome to do that. I'm just saying what I'd
like our *priorities* to be as reviewers and developers.

Competent, Friendly Bugzilla and Perl Services. Everything Else, too.
To view or change your list settings, click here:

Gervase Markham

May 13, 2009, 4:47:50 AM5/13/09
On 07/05/09 23:44, Max Kanat-Alexander wrote:
> No new features for Bugzilla 3.6. Instead, we focus on fixing the
> features we already have.

Is there a tentative release time for 3.6? That is to say, how long do
you suggest the project is focussed only on fixes?

Will this policy have any effect on the acceptability of, or speed of
reviewing, external contributions from potential new community members?


0 new messages