BGP-4+

John W. Stewart III jstewart@metro.isi.edu
Thu, 19 Dec 1996 10:46:12 EST


 > The expansion of the AS/RD space is an orthogonal issue, and one that is
 > far more difficult to achieve in a backward compatible fashion.  I would
 > argue that this is beyond the scope of the hack on the table...

i understand what you're saying

but it makes me wonder what the goal is of these bgp4 extensions.
i had thought that it was mainly to make it easier for providers
to start routing ipv6:  by allowing bgp4 to do it, then we get to
use a protocol we already know very well.  by allowing bgp4 to
route ipv6 (as well as any other protocol with an Address Family
identifier) then couldn't bgp4 become even more entreanched than
it already is?  in other words, if bgp4 becomes multiprotocol,
then why would i *ever* care to switch to idrp?  and if the result
of this is an increased lifetime for bgp4, then shouldn't the
future of the AS space be addressed now?

if the authors of the draft had envisioned a different future,
then perhaps it would be useful for that to be spelled out

/jws

 > 
 >    X-Auth: NOLNET SENDMAIL AUTH
 >    Date: Wed, 18 Dec 1996 17:31:25 -0600 (CST)
 >    From: Brandon Black <photon@nol.net>
 >    Mime-Version: 1.0
 >    Content-Type: TEXT/PLAIN; charset=US-ASCII
 >    Sender: owner-idr@merit.edu
 >    Precedence: bulk
 > 
 >    On Wed, 18 Dec 1996, Yakov Rekhter wrote:
 > 
 >    > John,
 >    >  
 >    > >  > > also, from reading the draft, i'm assuming that you plan
 >    > >  > > to support only ASs and not RDs? did you consider typing
 >    > >  > > routing domains in addition to network addresses?
 >    > >  > 
 >    > >  > I am a bit confused - in my mind "RD" and "AS" describe
 >    > >  > pretty much similar thing. 
 >    > > 
 >    > > AS is two bytes, while RD is variable.  right? or is
 >    > > it me that's confused?
 >    > 
 >    > Ok, I got it (I was confused). 
 >    > 
 >    > To answer your question, it doesn't look like we'll have shortage of 
 >    > ASs any time soon. So, it is not clear if there is a need for variable
 >    > length RDs.
 >    > 
 >    > Yakov.
 >    > 
 > 
 >    Famous last words :)
 > 
 >    Bill Gates I think once said something along the lines of "nobody will
 >    _ever_ need more than 640 kilobytes of main memory"
 > 
 >    And somebody (some whole group of bodies) once thought that a 32-bit IP
 >    address would hold off for along time....
 > 
 >    Just food for thought...
 > 
 >    .................................             ..............
 >    : Brandon Lee Black  : [Office] :.............: [Personal] :....
 >    :....................: brandon.black@wcom.com : photon@nol.net :.......
 >    : "Sanity is the     : +1.281.362.6466 .......: photon@gnu.ai.mit.edu :
 >    : trademark of a     :.................:..../\: vis_blb@unx1.shsu.edu :
 >    : weak mind. . ."    : LDDS WorldCom, Inc. :\/: +1.281.397.3490 ......:
 >    :....................:.....................:..:.................:
 > 
 >