Request for comments: New Bugzilla-based sponsoring process

[Dalibor Topic-5]

Good morning everyone,

thanks to Mark and the contributors on this mailing list we now
have a new Bugzilla-based contribution process [1]. It's time
to define the Bugzilla-based sponsoring process, thereby
complementing the new contribution process.

I've posted a first draft [2] on our extremely easy to upload to
code-review server (thanks, Tim!).  Comments are very welcome.

cheers,
dalibor tpoic


[1] http://openjdk.java.net/contribute/
[2] http://cr.openjdk.java.net/~robilad/sponsoring.html
--
*******************************************************************
Dalibor Topic                   Tel: [phone]
Java F/OSS Ambassador           AIM: robiladonaim
Sun Microsystems GmbH           Mobile: [phone]
Nagelsweg 55                    http://openjdk.java.net
D-20097 Hamburg                 mailto:Dalibor.Topic@...
Sitz der Gesellschaft: Sonnenallee 1, D-85551 Kirchheim-Heimstetten
Amtsgericht München: HRB 161028
Geschäftsführer: Thomas Schröder, Wolfgang Engels, Dr. Roland Bömer
Vorsitzender des Aufsichtsrates: Martin Häring

[john.coomes]

Dalibor Topic (Dalibor.Topic@...) wrote:
> Good morning everyone,
>
> thanks to Mark and the contributors on this mailing list we now
> have a new Bugzilla-based contribution process [1]. It's time
> to define the Bugzilla-based sponsoring process, thereby
> complementing the new contribution process.
>
> I've posted a first draft [2] on our extremely easy to upload to
> code-review server (thanks, Tim!).  Comments are very welcome.

Hi Dalibor,

Looks good; I have a couple of comments.

First, section "2. Link the contribution back to a Sun bug" includes
this:

        The Sun bug's description can just be a hyperlink straight to
        the Bugzilla bug.

Using the URL as the description isn't very useful, since bug
descriptions show up in all kinds of reports and in searches of Sun's
internal bug database.  I think both bugs should have the same
description; the hyperlink to the bugzilla bug can be used as the Sun
bug evaluation.

Second, we include bug numbers in changeset comments.  Should we use
the Sun bug id, the mozilla bug id, both?

-John

[john.coomes]

I (John.Coomes@...) wrote:

> Dalibor Topic (Dalibor.Topic@...) wrote:
> > Good morning everyone,
> >
> > thanks to Mark and the contributors on this mailing list we now
> > have a new Bugzilla-based contribution process [1]. It's time
> > to define the Bugzilla-based sponsoring process, thereby
> > complementing the new contribution process.
> >
> > I've posted a first draft [2] on our extremely easy to upload to
> > code-review server (thanks, Tim!).  Comments are very welcome.
>
> Hi Dalibor,
>
> Looks good; I have a couple of comments.
>
> First, section "2. Link the contribution back to a Sun bug" includes
> this:
>
> The Sun bug's description can just be a hyperlink straight to
> the Bugzilla bug.
>
> Using the URL as the description isn't very useful, since bug
> descriptions show up in all kinds of reports and in searches of Sun's
> internal bug database.  I think both bugs should have the same
> description; the hyperlink to the bugzilla bug can be used as the Sun
> bug evaluation.

Uh, nevermind this first comment.  You wrote "description," but I read
it as "synopsis."  Using the URL for the description is fine.

> Second, we include bug numbers in changeset comments.  Should we use
> the Sun bug id, the mozilla bug id, both?

This I'd still like to know.

Finally, since it's a manual process, it's inevitable that some
mozilla & sun interal bug states will diverge.  It'd be nice if there
was a simple way to list bugs that are not in sync, by engineer and/or
by category.

-John

[Dalibor Topic-5]

John Coomes wrote:

> I (John.Coomes@...) wrote:
>> Dalibor Topic (Dalibor.Topic@...) wrote:
>>> Good morning everyone,
>>>
>>> thanks to Mark and the contributors on this mailing list we now
>>> have a new Bugzilla-based contribution process [1]. It's time
>>> to define the Bugzilla-based sponsoring process, thereby
>>> complementing the new contribution process.
>>>
>>> I've posted a first draft [2] on our extremely easy to upload to
>>> code-review server (thanks, Tim!).  Comments are very welcome.
>> Hi Dalibor,
>>
>> Looks good; I have a couple of comments.

Thanks, John!

>> Second, we include bug numbers in changeset comments.  Should we use
>> the Sun bug id, the mozilla bug id, both?

That's a very good point, I updated the proposal to say:

"The changeset comment should reference the Sun bug id."

The new version is at
http://cr.openjdk.java.net/~robilad/sponsoring.html%2c1

Rationale:

Why not both bug ids? Because they should be linked to each other through
mechanisms described in contribution and sponsoring process documents,
so it should be sufficient to link to one of them to find the other.

Why not the bugzilla id? A contribution needs to have an associated Sun
bug id, which either exists at the time of contribution being made into
Bugzilla, or needs to be created by the Sponsor. Since a contribution
has a Sun bug id by the time it's being integrated, it's more consistent
with existing practice to continue to use it in changeset comments, at
least in the first phase of Bugzilla use in OpenJDK.

> This I'd still like to know.
>
> Finally, since it's a manual process, it's inevitable that some
> mozilla & sun interal bug states will diverge.  It'd be nice if there
> was a simple way to list bugs that are not in sync, by engineer and/or
> by category.

Good point - I added bugzilla issue 100018 for that.

cheers,
dalibor topic
--
*******************************************************************
Dalibor Topic                   Tel: [phone]
Java F/OSS Ambassador           AIM: robiladonaim
Sun Microsystems GmbH           Mobile: [phone]
Nagelsweg 55                    http://openjdk.java.net
D-20097 Hamburg                 mailto:Dalibor.Topic@...
Sitz der Gesellschaft: Sonnenallee 1, D-85551 Kirchheim-Heimstetten
Amtsgericht München: HRB 161028
Geschäftsführer: Thomas Schröder, Wolfgang Engels, Dr. Roland Bömer
Vorsitzender des Aufsichtsrates: Martin Häring

[Dalibor Topic-5]

Dalibor Topic wrote:
> Good morning everyone,
>
> thanks to Mark and the contributors on this mailing list we now
> have a new Bugzilla-based contribution process [1]. It's time
> to define the Bugzilla-based sponsoring process, thereby
> complementing the new contribution process.
>
> I've posted a first draft [2] on our extremely easy to upload to
> code-review server (thanks, Tim!).  Comments are very welcome.


Thanks to everyone who participated in the review, and contributed
comments.

The sponsoring process document is now final. It is available
at http://openjdk.java.net/sponsor/ , and cross-linked with
the contribution process document at http://openjdk.java.net/contribute/ .

cheers,
dalibor topic
--
*******************************************************************
Dalibor Topic                   Tel: [phone]
Java F/OSS Ambassador           AIM: robiladonaim
Sun Microsystems GmbH           Mobile: [phone]
Nagelsweg 55                    http://openjdk.java.net
D-20097 Hamburg                 mailto:Dalibor.Topic@...
Sitz der Gesellschaft: Sonnenallee 1, D-85551 Kirchheim-Heimstetten
Amtsgericht München: HRB 161028
Geschäftsführer: Thomas Schröder, Wolfgang Engels, Dr. Roland Bömer
Vorsitzender des Aufsichtsrates: Martin Häring

[Brad Wetmore]

 From a while ago.  I don't think Mark/Dalibor gave an answer to this yet...

> Second, we include bug numbers in changeset comments.  Should we use
> the Sun bug id, the mozilla bug id, both?

For now, use the Sun Bug id.  In Dalibor's final sponsor document, he
said to create a Sun bug if one does not already exist, and link the
Bugzilla bug to the sun bug.

Once we start tracking all bugs in Bugzilla, not just contributions,
we'll probably change that.

Brad