[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ih] inter-network communication history

On Fri, Nov 8, 2019 at 12:41 PM Brian E Carpenter <
brian.e.carpenter at gmail.com> wrote:

> Marshall Rose once told me that he regretted the "S" in SNMP, because it
> was anything but simple. And I think he also said that ASN/1 was chosen
> mainly to build bridges with the OSI world. But SNMP was a success (about
> 300 related RFCs exist).

Re: ASN.1.  My recollection of the history runs as follows.  HEMS needed an
extensible external data format -- we wanted folks to be able to add MIB
extensions w/o having to publish them. That is,  you'd just retrieve a
module of variables associated with an interface and in addition to the
standard variables you'd get anything else the vendor wanted you to know.
And at the time, almost all external data formats were of the form
"receiver knows what is coming".  So XDR and NDR and such.  I didn't want
to invent an external data format so I asked around BBN and Debbie Deutsch,
who had been part of the team that developed the ASN.1 BER said "how about
ASN.1?".  It met the need, so I grabbed it.  ASN.1 also made the OSI folks
happy but that wasn't the goal.*

When Jeff Case and team decided NSFNET couldn't wait for HEMS, they grabbed
what they considered the best parts of HEMS and subsetted it to make SGMP
(which became SNMP).  Now the OSI folks were around and it may be that Jeff
and team decided that to avoid annoying the OSI crowd, they'd stick with
ASN.  I don't recall them saying that, but it is possible and would be
consistent with Marshall's remark.


*The interactions of HEMS and SNMP with OSI were politically complex.  But
technically it was pretty simple.  Within six weeks of the HEMS effort
starting we were ahead of the OSI network management efforts and we stayed
ahead until HEMS ended.  I would argue that SNMP was also ahead as soon as
SNMP was implemented.  The OSI thinking at the time was fragmentary (not
enough folks were trying to manage a big network) and, if I remember right,
the first time the OSI network management standards went up for a vote they

Craig Partridge's email account for professional society activities and
mailing lists.
Internet-history mailing list
Internet-history at elists.isoc.org