Is It Worth Installing F9 Alpha?

[Michael Schwendt]

On Mon, 10 Mar 2008 04:56:16 -0700, Andrew Farris wrote:

> > The recent F8 kernel update is in the same area. In bodhi it's at karma -6
> > already, not counting anonymous users. The first tester there gave it +1
> > although he had to delete/reconfig his network profiles (which probably
> > was the same bug that hit me and killed the network).
> 
> And then noone else bothered during the entire week it was in updates-testing. 
> A lack of manpower testing or just people too lazy to help provide feedback to 
> the *community*-driven distro...

That's a great attitude. Not. That kernel update was forced upon the
community. It will happen again. In the hope to find more guinea-pigs. It
was not even waited for at least a sufficient number of positive karma
points. No real interest to be careful with such an upgrade. The warning
in the first comment was ignored. Later, negative karma and a request to
not push this update were ignored deliberately, too, because the people who
wanted to rush with this update have the freedom to overrule anyone else.

> Its not exactly hard to do on bodhi, so can 
> making testing easier really be where the blame lies? 

Give more guarantees about what impact karma points in bodhi have.
Recently, Josh Boyer mentioned that a -3 rating would block an update
automatically. I dunno where this has been mentioned before, but it is
very important information. Make sure that negative karma points spent
after a push request are not ignored either.

> Hardly anyone took notice 
> of that kernel until it hit updates,

That's the problem of such a flood of updates. The bodhi newsfeed here
only displays 20 items. The updates-testing report is very long and
not very readable.

> then all of a sudden there is feedback. 
> Guess what?  That pretty much means people weren't actually testing.

Hear, hear! You want community-driven kernel updates? Then do what has
been suggested before. Release a kernel only after a minimum number of
positive votes from within the community. Make it more transparent who
decides whether to unleash an untested kernel update upon the community.