FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Redhat > Fedora Development

 
 
LinkBack Thread Tools
 
Old 03-13-2010, 03:09 AM
Stephen John Smoogen
 
Default RFC: Bodhi voting method.

As has been noted by several people, the current voting method has
some short comings on what should be voted -1, 0, or +1. In order to
help clarify what to vote, and when here are some guidelines that
should be useful.

Vote Non-exhaustive Reasons for vote
================================================== ===================
-1 Program (or programs) do not work as expected.
Programs that link or require updated package do not work.
Bugs said to be fixed are not fixed.
Abrt reports a problem.
Installation/removal of RPM reports problems with scriplets, pre,
or post parts.
RPM spec changes have broken fedora guidelines.
---------------------------------------------------------------------
0 Program works as expected
Programs that link or require updated package work as expected.
Abrt does not indicate a problem after using.
Upgrade/installation does not indicate problems with rpm, spec
file, etc.
---------------------------------------------------------------------
+1 All of 0 reasons
Bugs known to user or bugs listed in reason for updated have been
tested and fixed (indicate in comments which bugs and test case
used.)


So for most users, when updating to updates-testing (or something
straight from bodhi/koji that has not been put in updates-testing) a 0
is the most likely response that should be given. A +1 should only be
given in cases where a specific test has been done (and should be
commented on. Ones without comments should be disregarded before pushing
to stable.)

As with any guidelines, this does not cover corner cases, emergencies,
or nitpicking to just find every problem possible. However it should
cover enough to make voting more useful.

If accepted, this may affect proposals that say that a certain number
of +1's are required before being pushed to production, but only after
testing and evaluation.



--
Stephen J Smoogen.

Ah, but a man's reach should exceed his grasp. Or what's a heaven for?
-- Robert Browning
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
 
Old 03-13-2010, 07:40 AM
Adam Williamson
 
Default RFC: Bodhi voting method.

On Fri, 2010-03-12 at 21:09 -0700, Stephen John Smoogen wrote:

> So for most users, when updating to updates-testing (or something
> straight from bodhi/koji that has not been put in updates-testing) a 0
> is the most likely response that should be given. A +1 should only be
> given in cases where a specific test has been done (and should be
> commented on. Ones without comments should be disregarded before pushing
> to stable.)

It's a nice effort, thanks. This doesn't reflect how we've been using it
in practice, though, certainly for F13. We're basically using it as a
quick 'sanity check'; Bill and I have been happily +1ing any critpath
package in F13 updates-testing which didn't actually eat our babies,
even if it has problems.

I think before we can get down to the nitty-gritty of fleshing out
specific voting criteria, we should probably agree on exactly what
extent we want testing to be done. And possibly whether we can really
represent it with a simple numeric system...
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net

--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
 
Old 03-13-2010, 02:42 PM
Stephen John Smoogen
 
Default RFC: Bodhi voting method.

On Sat, Mar 13, 2010 at 1:40 AM, Adam Williamson <awilliam@redhat.com> wrote:
> On Fri, 2010-03-12 at 21:09 -0700, Stephen John Smoogen wrote:
>
>> So for most users, when updating to updates-testing (or something
>> straight from bodhi/koji that has not been put in updates-testing) a 0
>> is the most likely response that should be given. A +1 should only be
>> given in cases where a specific test has been done (and should be
>> commented on. Ones without comments should be disregarded before pushing
>> to stable.)
>
> It's a nice effort, thanks. This doesn't reflect how we've been using it
> in practice, though, certainly for F13. We're basically using it as a
> quick 'sanity check'; Bill and I have been happily +1ing any critpath
> package in F13 updates-testing which didn't actually eat our babies,
> even if it has problems.

No problem. I figured it would be a good way to at least give people
who are using Till's scripts to know what to say when. If we get a lot
of 0's it still means its being looked at which fits into the criteria
people were talking about elsewhere.

> I think before we can get down to the nitty-gritty of fleshing out
> specific voting criteria, we should probably agree on exactly what
> extent we want testing to be done. And possibly whether we can really
> represent it with a simple numeric system...

I agree. Its more complicated and trinary logic.

> Adam Williamson
> Fedora QA Community Monkey
> IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
> http://www.happyassassin.net
>
> --
> devel mailing list
> devel@lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/devel
>



--
Stephen J Smoogen.

Ah, but a man's reach should exceed his grasp. Or what's a heaven for?
-- Robert Browning
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
 
Old 03-14-2010, 08:36 PM
"Richard W.M. Jones"
 
Default RFC: Bodhi voting method.

On Fri, Mar 12, 2010 at 09:09:16PM -0700, Stephen John Smoogen wrote:
> As has been noted by several people, the current voting method has
> some short comings on what should be voted -1, 0, or +1. In order to
> help clarify what to vote, and when here are some guidelines that
> should be useful.

The problem is you're assuming that the Bodhi karma system is the way
it should be done. But I think there should be separate buttons for
each of the things you have mentioned:

> Program (or programs) do not work as expected.
> Programs that link or require updated package do not work.
> Bugs said to be fixed are not fixed.
> Abrt reports a problem.
> Installation/removal of RPM reports problems with scriplets, pre,
> or post parts.
> RPM spec changes have broken fedora guidelines.

> Program works as expected
> Programs that link or require updated package work as expected.
> Abrt does not indicate a problem after using.
> Upgrade/installation does not indicate problems with rpm, spec
> file, etc.

> Bugs known to user or bugs listed in reason for updated have been
> tested and fixed (indicate in comments which bugs and test case
> used.)

I also think that if an update is broken because of the fault of
another package (as with ntfs-3g / libguestfs here: [1]), then the
right thing to do is to file a bug with the other package.

Rich.

[1] http://admin.fedoraproject.org/updates/ntfs-3g-2010.3.6-1.fc12

--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming blog: http://rwmj.wordpress.com
Fedora now supports 80 OCaml packages (the OPEN alternative to F#)
http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
 
Old 03-15-2010, 12:11 AM
Stephen John Smoogen
 
Default RFC: Bodhi voting method.

On Sun, Mar 14, 2010 at 3:36 PM, Richard W.M. Jones <rjones@redhat.com> wrote:
> On Fri, Mar 12, 2010 at 09:09:16PM -0700, Stephen John Smoogen wrote:
>> As has been noted by several people, the current voting method has
>> some short comings on what should be voted -1, 0, or +1. In order to
>> help clarify what to vote, and when here are some guidelines that
>> should be useful.
>
> The problem is you're assuming that the Bodhi karma system is the way
> it should be done. *But I think there should be separate buttons for
> each of the things you have mentioned:

Actually I do not believe I am assuming that at all. My assumption were:

1) I have an TUI/GUI interface setup as it is currently.
2) I have a tool (Till's updates-test script).
3) I have no guidelines on how to use my one widget.

Personally I don't care if it becomes improved or ripped out and
replaced with a "Electroshock the package owner" button. The RFC is
only meant to cover what is available at this moment.



--
Stephen J Smoogen.

Ah, but a man's reach should exceed his grasp. Or what's a heaven for?
-- Robert Browning
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
 

Thread Tools




All times are GMT. The time now is 06:55 PM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org