Re: WG Last Call draft-ietf-sming-reqs-04.txt

[rpresuhn-lists@dorothy.bmc.com]

Hi -

This discussion of what role any of the versions of ASN.1
might play in the sming effort reminds of the debates we had
a dozen years ago when GDMO and SMIv1 were being defined.

Both communities recognized that ASN.1, as a data definition
language, was not sufficient, because what was needed was an
object definition language, which defined interfaces which
through data definitions could be mapped into protocol.

Both communities used ASN.1 for the definition of attribute
syntaxes, range constraints, and type definitions.

They diverged in how they went about defining the object
definition language itself.  The SNMP community tried to use
the now-obsolete ASN.1 macro notation.  This was probably
a mistake, and was corrected in the definition of SMIv2.
The CMIP community didn't use the ASN.1 macro notation to
define GDMO, but instead defined a distinct language for object
definitions, which permitted reference to ASN.1 definitions
for attribute syntaxes and the like.

The way in which SNMP (ab)uses OBJECT IDENTIFIERs is far
removed from their envisioned purpose in ASN.1  I think it's
reasonable to use ASN.1 for syntax and type definitions,
but that trying to use it as an IDL or object definition
language is just too much of a stretch.  If the SMI were just
a data definition language, ASN.1 would be a reasonable and
complete answer.  However, the SMI is NOT just data definition,
and even modern ASN.2 with its "information objects" is not
at the appropriate level of abstraction for specifying the
behaviour of managed systems.

At the same time, we need to be careful that we don't lose all
connection to ASN.1, particularly since our encoding rules and
transport mappings (e.g. RFC 1906 and its imminent successor)
are closely tied to ASN.1 and BER.

 ------------------------------------------------------
 Randy Presuhn          BMC Software, Inc.  1-3141
 randy_presuhn@bmc.com  2141 North First Street
 Tel: +1 408 546-1006   San José, California 95131  USA
 ------------------------------------------------------
 My opinions and BMC's are independent variables.
 ------------------------------------------------------