Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!ucsd!ucbvax!WARTHOG.NCSL.NIST.GOV!dougm
From: dougm@WARTHOG.NCSL.NIST.GOV (Doug Montgomery)
Newsgroups: comp.protocols.iso
Subject: Re:  binary abstract NSAP address length
Message-ID: <9010231720.AA13083@warthog.ncsl.nist.gov>
Date: 23 Oct 90 17:20:41 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: National Institute of Standards and Technology (NIST)
Lines: 30

>From: REIJS%SURFNET.NL@cunyvm.cuny.edu
>Subject: binary abstract NSAP address length
>To: iso@nic.ddn.mil
>
>Is the decimal encoding still valid (it is only not PREFFERRED)? I thought
>that at this moment e.g. X.25 does use decimal encoding. Correct?
>
>So in case there is a possibility that different encoded subnetworks are
>connected, then the gateways/interworking units must be able to translate
>from binary to decimal encodings. In that case it must be possible to
>translate a binary encoded address into a decimal encoded address en vice
>versa. How is it then possible to change the length of the binary abstract
>syntax? If it is 20 octets long, it does not fit in the decimal encoding!
>
>Can somebody explain this to me?

YES, the decimal encoding is still valid.  What is important here is
how Network protocols convey OSI NSAPs.  Protocols standards dictate
how OSI Network Addresses are encoded and coveyed in PDUs (so called
Network Protocol Address Information).  It is my understanding that
current use of X.25 to support the CONS conveys OSI NSAPs using the
preferred binary encoding.


In situations where it is required to interwork with protocols and
subnetworks that employ decimal encoding of *full OSI NSAPs* to do
addressing, then one may need to be careful about the length of NSAPs
with binary abstract syntax DSPs.

dougm