Re: [eman] [MIB-DOCTORS] mib doctor review of draft-ietf-eman-rfc4133bis-05.txt

[Benoit Claise <[e-mail]>]

Dan,
> On Tue, Jan 15, 2013 at 01:07:43PM +0000, Romascanu, Dan (Dan) wrote:
>   
>>>>> f) In 2.12.2, can we find a better phrase than "SNMP ARCH [RFC3411]
>>>>>     method of context identification"? For example, "an SnmpEngineID
>>>>>     and ContextName pair [RFC3411] for context identification".
>>>>>
>>>> [[DR]] This is the verbiage used in RFC 4133. Unless we need to fix a
>>> bug I suggest to avoid such changes.
>>>
>>> It is document clarity. But I won't insist on this, I just find the
>>> current text unnecessarily confusing.
>>>
>>>>> g) Overall, replace MIBs with MIB modules.
>>>>>
>>>> [[DR]] Same as the previous point.
>>> Again, it is document clarity. It is rather obvious that this change
>>> does not break anything, no? Let Bert decide - he is the owner of the
>>> one MIB many MIB modules song. ;-)
>>>
>> [[DR]] I also sing the song when I review new documents, but in this case we would interfere in the text of the document that is now at version 4.
> So what? Time to fix this. Anyway, I will not fight for this if
> editors believe it is best to not update vocabularity to current
I agree with Juergen on the two points above.

Every single time I speak SNMP/MIB with someone, I try to use the right 
language, and correct my interlocutor if needed, with a little of education.
I don't understand why we don't jump on the opportunity to correct this 
document.
There are two types of audience for this new document:
1. The people who have not read the first 3 versions. So I guess not 
that familiar with the SNMP/MIB. They would benefit from the right 
terminology
2. The persons who need a diff with the previous version. So I guess 
familiar with SNMP/MIB. And the improvements would be obvious to them.

Therefore, there is no harm in improving the document. And I don't buy 
in the argument that we need to keep consistency with RFC 4133 ..; for 
something that is not clear.

Regards, Benoit
> rules.
>
> /js
>