Re: guidelines and SMIv1 -> SMIv2 conversions

[Andy Bierman <abierman@cisco.com>]

At 06:51 PM 2/11/2003 -0800, C. M. Heard wrote:
>[ note change of subject line ]
>
>On Tue, 11 Feb 2003, Andy Bierman wrote:
>> I know I mentioned this before, but sometimes a lot of
>> guidelines and rules have to ignored when the MIB under
>> review is an SMIv1 to SMIv2 conversion.  I know that
>> doesn't come up in IETF reviews very often, but it
>> still happens from time to time in Cisco reviews. 
>> 
>> Since we (at Cisco) plan to enforce the guidelines
>> in this document for our Enterprise MIB reviews, I
>> think we should account for SMIv1 -> SMIv2 conversions.
>
>I certainly don't mind putting a note into the introduction that
>saying that some of the guidelines cannot be followed in such
>conversions, but I'm reluctant to point each specific case.
>
>Here is the proposed text of a note that to go at the end of
>Section 1 (in its own paragraph after the other additions
>discussed earlier today):
>
>   Reviewers and authors need to be aware that it will not be possible
>   to apply all of the guidelines in this memo to MIB modules that are
>   being translated to SMIv2 from SMIv1 (STD 16) [RFC1155] [RFC1212]
>   [RFC1215].
>
>The reason for my reluctance to do more than this is simply the
>amount of work involved.  However, if you or anyone else that
>would like to have the specific cases of guidelines that have to
>be ignored pointed out, I'm certainly not averse to incorporating
>contributed text.  Collecting the details into an appendix might
>be the easiest way to do that, but adding notes at strategic
>places in the body of the document would be OK as long as there
>aren't too many of them.
>
>Comments?

This is fine. I suppose it would be too hard to list all the details,
and misleading to list a couple.  The most common issue is that
SMIv1 INDEX components are read-only and in SMIv2 they are
supposed to be not-accessible.


>Mike

Andy