[Openstack-docs] Workflow for specs

Gauvain Pocentek gauvain.pocentek at objectif-libre.com
Thu Jul 31 19:34:07 UTC 2014


Le 2014-07-31 21:25, Anne Gentle a écrit :
> On Thu, Jul 31, 2014 at 2:04 PM, Andreas Jaeger <aj at suse.com> wrote:
> 
>> On 07/30/2014 09:43 PM, Gauvain Pocentek wrote:
>>> Hi all,
>>> 
>>> We have to define a workflow to define and approve specs. Anne 
>>> proposed
>>> the following on the HOT spec:
>>> 
>>> - approve the blueprint in Launchpad
>>> - then ask only some for docs-specs
>>> - then docs-specs is just used to ensure the approach and scope are 
>>> what
>>> we want
>>> 
>>> Anne please correct me if I'm twisting your words :)
>>> 
>>> My opinion is that we should create the Launchpad BP only after a 
>>> patch
>>> to docs-specs has been accepted. This would make sure that there's 
>>> been
>>> some thinking about how things can be done, and that we agree on the
>>> choices.
>> 
>> Do we have a reference what other projects are doing? I'd like to use
>> the same workflow as much as possible.
> 
> Ah, yes, I looked this up yesterday. 
> 
> Nova has a blueprint in Launchpad first, then marks it "Pending
> Approval" for Design. 
> 
>  Then once the nova-specs patch is accepted, the Design is marked 
> "Approved."

This does make sense. I'm all for it.

> 
> I think this makes sense to adopt for docs. Just a few caveats: nova
> has a release manager and docs does not. So the responsibility falls
> on the PTL to clickety-click in Launchpad so people know what will be
> delivered in a given release. 
> 
> I do think that we need to keep the component of very limited reasons
> to have a blueprint and upfront design at all. For nearly everything,
> we need writing more than designing the writing. 

Agreed!

Thanks Anne.
Gauvain


>> I have no strong reference either way.
>> 
>>> My experience with the HOT spec shows that there's a lot of 
>>> questions
>>> that need to be answered, and LP doesn't provide a simple way to
>>> discuss, modify, or keep the history of the spec definition --- 
>>> which I
>>> think is important. And we're used to git and gerrit anyway :)
>>> 
>>> The HOT spec review for reference: 
>>> https://review.openstack.org/#/c/108133/ [1]
>>> 
>>> What so you think?
>> 
>> Andreas
>> --
>>  Andreas Jaeger aj@{suse.com [2],opensuse.org [3]} Twitter/Identica: 
>> jaegerandi
>>   SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
>>    GF: Jeff Hawn,Jennifer Guild,Felix Imendörffer,HRB16746 (AG 
>> Nürnberg)
>>     GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 
>> A126
>> 
>> _______________________________________________
>> Openstack-docs mailing list
>> Openstack-docs at lists.openstack.org
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-docs 
>> [4]
> 
> 
> 
> Links:
> ------
> [1] https://review.openstack.org/#/c/108133/
> [2] http://suse.com
> [3] http://opensuse.org
> [4] http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-docs



More information about the Openstack-docs mailing list