[AFS3-std] Using Gerrit for draft review

[Douglas E. Engert]

On 12/7/2010 2:32 PM, Simon Wilkinson wrote:
>
> On 7 Dec 2010, at 19:14, Jeffrey Altman wrote:
>
>> Is this something the group would like to try?
>
> I think there are two issues
>
> The first is document format. If we're interested in reviewing differences between revisions of documents, then we would need to agree on a single, structured, submission format such as XML. This is
> different from that used by the I-D archive, and would penalise those who aren't prepared to go through the trauma of xml2rfc. Gerrit wouldn't be able to handle plain text.
>
> The second is whether gerrit is actually a good place to have wide ranging design discussions. Our experience with the OpenAFS gerrit suggests that there is a point where these kinds of conversations
> become unmanageable in gerrit, and have to migrate to email. This means that you end up with your history in two places, something I think would be a hinderance when it comes to writing protocol
> documents.

I tend to agree with Simon. We have a small set of developers and other
interested parties, who may not use gerrit. Requiring them or newcomers
to use it may not be as open as we would like.

>
> Cheers,
>
> Simon.
>
>
> _______________________________________________
> AFS3-standardization mailing list
> AFS3-standardization@openafs.org
> http://michigan-openafs-lists.central.org/mailman/listinfo/afs3-standardization
>
>

-- 

  Douglas E. Engert  <DEEngert@anl.gov>
  Argonne National Laboratory
  9700 South Cass Avenue
  Argonne, Illinois  60439
  (630) 252-5444