and to clear up a little confusion in the references in the previous

the OSI reference:
http://www.garlic.com/~lynn/2003m.html#9 OSI not quite dead yet

was in a thread about OSI/ISO vis-a-vis IETF ... using DNS activity as
example of IETF workings.

The thread started by claiming that two major differences between ISO and
IETF vis-a-vis OSI ... were

1) ISO disn't require actual implementation for something to become a
standard (you can get standards that are not actually implementable) while
IETF does

2) ISO attempted to cast OSI in concret and make a cult of it for the true
bleievers by instructing ISO-chartered standards body that they couldn't
standardize a networking protocol that violated the OSI model

http://www.garlic.com/~lynn/2003l.html#47 OSI not quite dead yet

Basically OSI model reflected 1960s era technology. To some extent the
ARPANET during the 70s reflected the OSI model. By the early '80s (when OSI
was finally passed, at least) two significant things had happened:

1) arpanet on 1/1/83 converted to internet(working) protocol ... something
that doesn't exist in the OSI model

2) LANs appeared that subsume levels 1, 2, and part of 3 (in the OSI model)
with a MAC interface that sits in the middle of layer 3.

Trying to get HSP (high-speed protocol) considered as standard in the late
'80s in x3.s3.3 (iso chartered us body responsible for networking protocols
in OSI level 3&4) ....  it had to be turned down. HSP would go directly
from level 4 interface directly to LAN/MAC interface. Since LAN/MAC
violated OSI ... and since ISO proscribed work on standards that violated
OSI model .... it was not possible to standardize any protocol that
supported LAN/MAC interface.

--
Internet trivia, 20th anv: http://www.garlic.com/~lynn/rfcietff.htm


Reply via email to