[rfc-i] Comments in the canonical RFCs
pkyzivat at alum.mit.edu
Mon May 12 11:19:12 PDT 2014
On 5/12/14 1:23 PM, Paul Hoffman wrote:
> On May 12, 2014, at 9:52 AM, Heather Flanagan (RFC Series Editor) <rse at rfc-editor.org> wrote:
>> On 5/9/14, 11:34 AM, Joe Hildebrand (jhildebr) wrote:
>>> +1. We probably want to retain the last edited copy of the XML separate
>>> from the published canonical version for other reasons (e.g. expanding
>>> TOC), and I don't mind that copy having all the comments left in. When
>>> you go to -bis that document, you can start from the last-edited version,
>>> rather than the canonical one.
>>> On 5/9/14, 1:08 PM, "Donald Eastlake" <d3e3e3 at gmail.com> wrote:
>>>> I support removal of comments before publication. Hidden potential
>>>> junk is something we don't want.
>> And this basically covers my concern and intent regarding comments in
>> the canonical version of an RFC. On the balance, comments within the
>> canonical file cause more problems than they solve. If someone needs to
>> comment on the XML used, WG or author notes to future -bis authors, or
>> offer some other minor clarifying point, I think that correctly belongs
>> in the draft, not the final RFC.
> When you say "if someone ... in the draft", that indicates that no comments are ever added after the final draft. However, any of us who has authored an RFC knows that is not the case: the RFC Editor team adds comments to the XML during the editing process. Given that, I think it is important that both the final draft *and* some representation of the comments that were added during the final editing process be retained for the author of the revision to the RFC.
> The easiest way to do this is to have a real VCS for the editing phase, something for which many of us have been wishing for quite some time. (This would have the side-effect of more predictable communications with authors during AUTH48.) Short of that, there needs to be a way to capture for posterity the marked-up XML that started the AUTH48 process.
That *concluded* the AUTH48 process!
More information about the rfc-interest