The following text is copyright 1994 by
Network World, permission is hearby given for reproduction, as long as
attribution is given and this notice is included.
By: Scott Bradner
The U.S. Federal
Internetworking Requirements Panel (FIRP) has issued a draft of its
reevaluation of the U.S. government's "requirements for open systems
networks and to recommend policy on the Government's use of networking
standards." I expect that by the time this column appears there will have
been a number of news stories about the draft and its recommendations. Since it
refers to some issues that I've written about in these columns in the past, I'd
like to make some more comments now.
This panel was convened
to examine the disconcerting (to some) fact that, although back in 1988 the
U.S. Government had defined a subset of the OSI networking standards, known as
GOSIP, as the official network protocols for all governmental use, (sort of
like the official toothpaste for the Olympic games, but in this case the
payment is reversed), the use of other network protocols, most notably TCP/IP,
continued to expand.
In a previous column (May 31 1993, page 21) I lamented that as far as
higher level corporate management, and by implication, governmental management
(if management be a legitimate word in this context) was concerned, the IETF
standards could garner little respect. This is in spite of the careful
standards development process that I described in a subsequent column (July 5, page 13).
There are positive many
things in this 40 page draft from the FIRP, but one of the most satisfying to
me as an Internet Engineering Task Force (IETF) attendee, an Internet
Engineering Steering Group (IESG) member, an Internet Society member and an Internet Society trustee is the
recommendation that "for the U.S. Federal Government, IETF standards
should qualify as 'open, international, voluntary standards' and, therefore
should qualify for use on an equal basis with standards from internationally recognized
standards organizations."
The board has determined
that the IETF
standards process and the resulting standards are legitimate and not the equivalent of
as a slightly off-color joke. This draft suggests that it is better to strive
for effective and cost efficient solutions to problems rather than to blindly
follow the holy grail of any particular set of standards. If these
recommendations are adopted then TCP/IP and OSI products will be not only
competing between themselves but also with IPX, SNA and AppleTalk to provide
the best match to the needs of the particular governmental organization.
This only applies to the
U.S. government and only if the draft is not modified during the comment
period. It does not mandate any changes in U.S. industry but, just maybe, the
action will be noted in some of the higher management layers. The people on the
lower levels, who have had the job of actually getting things to work have been
voting with their designs for quite a while.
This also only applies
within the U.S.. There are many other governments across the globe that have
established GOSIP-like requirements of their own. This U.S. document might be
used as an example of the kind of change that is best for the entire networking
community.
The text of the draft report is available via anonymous ftp from the
prophetically named computer "osi.ncsl.nist.gov". Comments on the
draft can be sent to the email address 'firp-comments@osi.ncsl.nist.gov. It is
also real nice to see that they prefer comments in electronic form instead of
paper.
-----------
Sign seen in the window
of a local restaurant: "all new breakfast - $2.99" - I don't want to
know what the alternative is.
Disclaimer: Opinions?
Harvard does not have opinions, it has dictates, and these ain't them
sob@harvard.edu