On 06/06/2010 12:33 AM, Geoff Beasley wrote:
On 06/06/2010 04:58 AM, Patrick Shirkey wrote:
>
> I like this idea and I can see a place for it at
Linuxaudio.org. A
> centralised feature/bug/infrastructure tracker.
If someone wants to step forward, hosting it under the umbrella (and on
the server) of
linuxaudio.org will be the least issue.
> I think the hardest part will be to isolate
the most important
> information
> and present it in a very obvious way. This could easily get left high
> and
> dry by making the system too complex for the info that is being
> aggregated.
Bug-tracking always requires user & developer interaction (confirm,
reproduce, comment..). Merly aggregating or collecting information won't
do much good.
The feeds provide a link through to the original post so that provides a
way back for contributors and developers.
OTOH I really like the idea to improve
interoperability between apps.
> A way to start the system could be to collate
the info using rss feeds
> from the various bug trackers that are already in use.
Testing this idea:
http://planet.linuxaudio.org/bugs/
currently collects [only] ardour, jack & qtractor tracker's feeds.
Imagine how this would look with over 50 projects. I doubt it would be
very useful. But things could be improved by using more elaborate
RSS/Atom feed queries..
Anyway, this is a one-way system. Users will need to use
upstream-trackers to submit information. This somehow undermines the
idea of providing feedback for interop issues at a central location.
The easiest way to supplement this would be a linux-audio-bugs
email-list or just re-use this list.
The bugs feeds is a good start. Did you just set that page up?
I don't like the idea of flooding this list with bug reports. A new list
is an option using the feeds from the bugs page as data. But that is
really just replicating existing functionality. A more productive approach
is to improve on the bugs page now that it exists. For example an "add" /
"submit" new feed button/link would be helpful.
linuxaudio.org
would indeed be a good place for such a tracker, and it
is indeed a good idea but for the issue of 'updatedness' ie; it would
need quite some effort to remain up-to-date and therefore relevant and
useful. If it slipped behind it could in fact have the opposite effect.
who would be prepared to undertake such a burden? it would need to be
completely automated as the amount of info could be quite substantial
quite quickly... seems to me the devil is in the detail.
Indeed. Implementing a centralized cross-project bug & feature tracker
- possibly with bounties - can become quite complex; not to mention it
requires quite some effort to maintain it.
Please proof me wrong. It'd be a great thing to have.
I think you are correct. The trick is figuring out how to move the concept
forward without getting bogged down in minor details.
--
Patrick Shirkey
Boost Hardware Ltd.