Updates lacking descriptions

[Adam Williamson]

On Thu, 2009-08-13 at 17:56 +0200, Michael Schwendt wrote:

> For me it isn't.  I won't spend extra time on writing special summaries
> for a test-update, if nobody contributes any testing. 

FWIW you almost certainly _are_ getting some testing. There are
definitely users on -test-list who run with -updates-testing enabled
permanently and hence run all updates-testing packages (that they have
installed, anyway). In most cases, however, they don't give positive
feedback when everything works, because the current mechanism makes it
too clumsy (go to bodhi, log in because it _will_ have forgotten you
again, find the update, type 'yay it works!' in the box, submit, rinse
and repeat for the next twelve updates). That's a process problem, we
need to make it easier to give a simple 'thumbs up' feedback. Where
things are broken, negative feedback usually does come through (whenever
someone sends something with broken dependencies to updates-testing, for
instance, someone yells about it on test-list and in bodhi quite
quickly, most of the time).

> Once the update has
> been marked stable, it's too late. It will be installed by some users
> "blindly". 

Sure, some, but not all. It doesn't have to benefit everyone to be a
good idea, just a decent amount of people.

> They won't base any decision on reading the update
> description. Not even the list of affected bugzilla tickets in bodhi
> implies that the fixes are correct or won't cause side-effects. A
> minor feature addition might cause the software to crash in untested
> environments. 

True.

> The update description doesn't add any quality.

False. Don't make the perfect the enemy of the good. The fact that
update descriptions are 'fallible' in the ways you describe does not
make them useless.

-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net