Brian E Carpenter <[EMAIL PROTECTED]> writes:
> Actually the former FP was clearly a variable length object. Anyway,
> I read you to mean that chopping out a large fraction of the global
> unicast space for son-of-8+8 is an option left open by this
> architecture.
Yes. If someone comes along and makes a case that some chunk of
address space needs to be reserved for a specific purpose other than
global unicast addresses, that request would be evaluated on its
merits when it is made.
> That's fine, as long as people realise that such addresses would
> contain mutable bits, with impact on checksum algorithms.
Perhaps. :-) I.e., any proposal to reserve some swath of address space
for some new purpose would presumably also explain how the space will
be used, how nodes implementing the new service interact with devices
that don't understand the new usage (if they need to), etc.
Thomas
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
