Re: Naming Conventions for Descriptors (was: comments/review<draft-ietf-ops-mib-review-guidelines-00.txt>)

[Juergen Schoenwaelder <schoenw@ibr.cs.tu-bs.de>]

>>>>> C M Heard writes:

[...]

Mike> I'm not sure how to fix this or even whether you are actually
Mike> asking me to do so.  So, I have not (yet) changed anything in
Mike> response to this comment.

I think we should not go overboard with our guidelines and I am thus
totally happy to do nothing here. ;-)

Mike> OK, I have reworked the proposal to use my favorite example.
Mike> Unfortunately that is still an Internet-Draft (albeit one that
Mike> is in IETF last call).  If you have an equally good example that
Mike> is already published as an RFC let me know and I will use it
Mike> instead.

Equally good will be hard to measure. I personally would probably use
something like the ENTITY-MIB which models something which is less
specific and more generally applicable (and thus the MIB is probably
wider known).

Looking at the OPT-IF-MIB, I actually tend to dislike the module
naming convention. I believe that if should be called IF-OPT-MIB
rather than OPT-IF-MIB since it actually seems to be an extension of
the IF-MIB for optical interfaces. I generally prefer if the most
general part of a module name is first. Example MIBs that follow
this rule are:

APPN-*
OSPF-*
ADLS-*
ATM-*
DNS-*
DOCS-*
ENTITY-*
RADIUS-*
PPP-*
IF-*
IP-*
IPV6-* (although they are being replaced)
SNMP-*
DISMAN-*

plus probably a few others. There are of course also examples that do
not follow this style such as *BRIDGE-* MIBs and the RMON MIBs. But
the majority is I think following the style which I prefer.

/js

-- 
Juergen Schoenwaelder    <http://www.informatik.uni-osnabrueck.de/schoenw/>