I like this. In particular I like the clean use of “version” and “revision”. 
Editorial nit: add a comma after “i.e.” because that’s the style used for 
“e.g.”. Tx, W.

> On 31 Aug 2016, at 11:56, Jonathan Hansford <[email protected]> wrote:
> 
> How about:
>  
> NEW:
>  
> It is not required to keep the full revision history of draft versions (e.g., 
> modules contained within Internet-Drafts). That is, within a sequence of 
> draft versions, only the most recent revision need be recorded in the module. 
> However, whenever a new (i.e. changed) version is made available (e.g., via a 
> new version of an Internet-Draft), the revision date of that new version MUST 
> be updated to a date later than that of the previous version.
>  
> Jonathan
>  
> From: William Lupton <mailto:[email protected]>
> Sent: 29 August 2016 15:20
> To: Andy Bierman <mailto:[email protected]>
> Cc: [email protected] <mailto:[email protected]>
> Subject: Re: [netmod] RFC 6087bis guidance re use of revision statements 
> indrafts
>  
> Andy,
>  
> This thread started with discussion of an apparent ambiguity in the current 
> text:
>  
> OLD
>  
> It is acceptable to reuse the same revision statement within unpublished 
> versions (i.e., Internet-Drafts), but the revision date MUST be updated to a 
> higher value each time the Internet-Draft is re-posted.
>  
> —— 
>  
> It became clear from the subsequent discussion (thanks Randy!) that the above 
> text isn’t intended to mean “reuse the identical revision statement, 
> INCLUDING THE REVISION DATE” but to mean “reuse the revision statement, 
> UPDATING THE REVISION DATE”.
>  
> Then other people raised other points, e.g only updating the revision date if 
> the YANG has changed, distinguishing between the document and the YANG 
> contained therein, and distinguishing between YANG in IDs and YANG created by 
> other SDOs. My proposed new text tries to address all of these:
>  
> NEW:
> 
> It is not required to keep the full revision history of draft versions (e.g., 
> modules contained within Internet-Drafts). That is, within a sequence of 
> draft versions, only the most recent revision need be recorded in the module. 
> However, if the module has changed, the revision date of the most recent 
> revision MUST be updated to a later date whenever a new version is made 
> available (e.g., via a new version of an Internet-Draft).
>  
> ——
>  
> I believe that this retains the original intent in a way that resolves the 
> original ambiguity and addresses the other points that were raised. It it’s 
> “worse”, how is it worse (apart from being longer, on which point mea culpa)?
>  
> Thanks,
> William
>  
> On 19 Aug 2016, at 15:42, Andy Bierman <[email protected] 
> <mailto:[email protected]>> wrote:
>  
> On Fri, Aug 19, 2016 at 7:13 AM, Dale R. Worley <[email protected] 
> <mailto:[email protected]>> wrote:
> Andy Bierman <[email protected] <mailto:[email protected]>> writes:
> > An Internet-Draft is NOT a means of "publishing" a specification;
> 
> As I said, that's the theory, but practice is considerably different.
>  
> Anybody that implements a work-in-progress knows they are taking a risk
> on an unstable document.  The guideline already says MUST update
> the revision date.
>  
> Not sure what more you want to guidelines document to do.
>  
> Dale
>  
> Andy

_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod

Reply via email to