launchpad-dev team mailing list archive

[Deryck Hodge <deryck.hodge@xxxxxxxxxxxxx>]

On Thu, Mar 3, 2011 at 5:01 PM, Gavin Panella
<gavin.panella@xxxxxxxxxxxxx> wrote:
> On 2 March 2011 18:29, Curtis Hovey <curtis.hovey@xxxxxxxxxxxxx> wrote:
> ...
>> I favour addressing the second issue to solve the oops reported by the
>> first issue. I think deleting bugtasks would be great, but it is hard to
>> justify with 250+ critical bugs. I believe the rules are more complex
>> than I stated.
>
> Preventing under-privileged users from re-targeting bugtasks in
> specific states may require extra work in terms of conditionally
> disabling pickers and forms, or, more simply, letting them try and
> then telling them it can't be done. Either way it's more rules for
> users to learn, or for us to attempt to make clear in the UI.
>
> Making bugtask targets immutable and allowing bugtask deletion will
> make the UI less noisy and easier to understand, will get rid of the
> need for the null project, will probably let us simplify a lot of
> code... with naive optimism I'd like to think that everything will be
> simpler.
>
> Do you know what the issues are with deleting bugtasks?
>
> Although I agree that the restriction approach might be the most
> expedient solution, I do think it's non trivial. It also feels a
> little like procrastination instead of fixing the "big" problem of
> bugtask deletion.
>

FWIW, I also think bugtask deletion is the right way forward.  It's
the real bug.  I'm not sure it's more work.  I think it's 6 in on
hand, half dozen in the other, as users of cliches say.  Adding
restrictions has hidden complexity and tends to lead to newer bugs
being filed.  So in the long run it's definitely more work.

Cheers,
deryck


-- 
Deryck Hodge
https://launchpad.net/~deryck
http://www.devurandom.org/