Re: [Gen-art] Gen-ART review of draft-ietf-pce-pcecp-interarea-reqs-05.txt

I think Miguel is completely correct in principle -
requirements shouldn't include specific protocol elements;
they should describe necessary functionality.

This draft has to fit into a rather precisely defined world
though, so I don't think it can really be made more abstract.
I won't ask for any changes.

      Brian

On 2007-03-05 13:42, Miguel Garcia wrote:
> I have been selected as the General Area Review Team (Gen-ART)
> reviewer for this draft (for background on Gen-ART, please see
> http://www.alvestrand.no/ietf/gen/art/gen-art-FAQ.html).
> 
> Please wait for direction from your document shepherd
> or AD before posting a new version of the draft.
> 
> Document: draft-ietf-pce-pcecp-interarea-reqs-05.txt
> Reviewer: Miguel Garcia <miguel.an.garcia@nokia.com>
> Review Date: 2007-03-05
> IESG Telechat date: 08 March 2007
> 
> Summary: The document is ready for publication as Informational RFC.
> 
> Comments: I have only a very subjective comment. In my opinion 
> requirements drafts should make an abstraction of the solution and just 
> note down the requirements. However, while reading this draft, I got the 
> impression that the authors have been thinking quite a lot on the 
> solution space, haven't separated them from the solution, and have 
> written down a few solutions. Just to illustrate a couple of examples. 
> On Section 4.8 the text reads:
> 
>    The request message MUST allow for the inclusion of the address of
>    the originating PCC.
> 
> This is a solution for a requirement. Unfortunately it is not clear to 
> me what the requirement is. It is probably something related to the 
> "ability of a PCE to apply PCC-specific policies" or something like 
> that, where a solution is "to record the address of the PCC in request 
> messages, so that the PCE can apply a pcc-specific policy".
> 
> Another example in Section 4.4:
> 
>    Hence the request message SHOULD allow a request for the
>    identification of path segments computed by a PCE, and the response
>    message SHOULD allow identifying the path segments computed by each
>    PCE.
> 
> Well, just an opinion.
> 
> /Miguel
> 
> 

_______________________________________________
Gen-art mailing list
Gen-art@ietf.org
https://www1.ietf.org/mailman/listinfo/gen-art