Backwards development

[mva at sysfault.org]

Quoting Gervase Markham <gerv at mozilla.org>:

> Hi Denis,
>
> On 18/11/15 20:16, Denis Roy wrote:
>> Why does Mozilla do this backwards?  Wouldn't you want to improve the
>> core project and have BMO (and everyone else) consume that?
>
> In the last couple of years, we have been struggling with the impedance
> mismatch of BMO's rapid release cycle (they ship weekly) and regular
> requirements for improvements from Mozilla's management, with the
> differing pace of development of upstream. For better or worse, the BMO
> team have decided that working upstream, doing releases and then
> shipping them on bugzilla.mozilla.org is not a model which works given
> their constraints.
>
> They do plan to merge the trunk into their codebase again soon, and I
> hope that the upstream developers (as many as there are) will be able to
> work together to take that opportunity to port upstream many of the
> generally-applicable improvements that have been made on BMO.
>
>> Short story -- does the Bugzilla project want contributions, or is it
>> relegated to being a code dump for Mozilla's bug tracker?
>
> The Bugzilla project is extremely keen to see contributions,
> particularly from outside Mozilla. In fact, Bugzilla's future as an
> independent software project depends on it.

It does not feel like that. Start to eat your own dog food :-).
Use a stock Bugzilla for tracking Bugzilla bugs instead of BMO - right now the
pressure on the project itself for its own needs is low, since it
benefits from things (BMO) that are not merged into it.

How about not closing bugs and feature requests as duplicates,  
referring to other
bugs being more that 4 or 5 years old (with the last comment being  
equally old)?
This neither leaves a good impression nor motivates others to participate even
more. Take care of them instead. I have quite a set of changes and  
enhancements,
which I am currently not discussing on the tracker, since I

a) won't get any response (or do not expect to get any, based on my  
experience)
b) can expect them to be closed with a referral to an aged bug and thus simply
    vanish
c) will be told that this is an incomplete solution, since it's not  
generic enough

Finally, push small changes and features into Bugzilla, even if they only
deal with a minor bit of a bigger solution. My impression right now is  
that lots
of things are hold back to aim for a "generic, one-size-fits-all" approach.
Although that's appealing, it will delay features and is not a motivator for
contributors. Implement a small feature, then enhance it and get  
feedback from users
about whether your approach is correct (learn from your failures, do  
not live with them).

Cheers
Marcus