Fate vs Bugzilla for feature tracking

[Ingo Klöcker]

On Wednesday 21 January 2009, Aaron J. Seigo wrote:
> hello =)
>
> after a few weeks of (patient :) consultation with the Suse team
> working on Fate, i just blogged about the possibility of using Fate
> for feature tracking in KDE. you can read about it here:
>
> 	http://aseigo.blogspot.com/2009/01/do-you-believe-in-fate.html
>
> so i won't bother repeating it all over again in this email.

I commented to your blog. For convenience, I'll repeat my comments here. 
Some of them have already been raised by Andras and Michael, still here 
they are:

I think having one website for bug reports and a different website for 
feature requests will be a nightmare for everybody (users, bug triage 
team, and developers). Many bugs filed on Bugzilla are actually feature 
requests (at least from the developers' point of view). Those would 
have to be cloned on Fate. Vice-versa I'm pretty sure many bugs would 
be filed on Fate instead of on Bugzilla. Often it is unclear (and 
impossible for the users to decide) whether something is a bug or a 
missing feature.

Either we settle on an alternative to Bugzilla for both, bugs and 
features, or we stay with Bugzilla and try to correct it's 
deficiencies.

Another alternative would be using Fate as tool for developers only, 
i.e. only developers would be able to create feature requests on Fate. 
If a developer wants to discuss (and eventually implement) a feature 
posted on Bugzilla he could create a new entry on Fate and link it to 
the corresponding Bugzilla entry. This way the developers could take 
advantage of Fate without confusing the users by having two different 
spots for reporting bugs resp. for submitting feature requests.


Regards,
Ingo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20090121/c4a608af/attachment.sig>