Re: Prior art on IPv6-in-IPv4 tunneling

[Brian E Carpenter <brian@hursley.ibm.com>]

Fred,

I'm not quite sure what point you are making here or why. That work was 
in fact the first implementation of 6over4 - the authors of RFC 2529 were 
in contact with Lixia Zhang and Quang Nguyen. The first publication was 
draft-carpenter-ipng-6over4-00.txt in November 1996, discussed at
the IPNGWG meeting in December 1996. 

I agree that Quang Nguyen's paper was interesting and thought around the
topic. But his code was 6over4 and the prior art for that is from 
November 1996. Both the authors of 6over4 have changed employers since
then, but to the best of my knowledge no IPR claims were made.

Of course the basic idea of tunnelling is older and trivial.

   Brian

"Fred L. Templin" wrote:
> 
> Dear friends,
> 
> It has very recently been brought to my attention that a defining work
> on IPv6-in-IPv4 tunneling exists dating back to Spring 1998 and entitled:
> 
>    "Virtual Ethernet: A New Approach to IPv6 Transition"
> 
> The work was prepared by a UCLA Masters Student named Quang Nguyen
> who was advised by Dr. Lixia Zhang. The document describes the case
> of intra-site tunneling when IPv4 multicast is deployed (as in 6over4)
> but also gives a detailed study of the case in which IPv4 multicast is
> absent (as in isatap).
> 
> Upon examining the document, I realized that concepts presented in
> the isatap specification match up very closely with the prior art so
> I contacted Dr. Zhang to arrange a meeting to discuss the similarities.
> Dr. Zhang and I were able to meet today, and she has requested that an
> acknowledgement to her student and a reference to the document be given
> in the specification. This action will be taken asap in the next isatap
> draft revision.
> 
> The document in question can be found at:
> 
>    http://irl.cs.ucla.edu/vet/report.ps
> 
> I would encourage other v6ops/ngtrans authors to examine this work for
> prior art. For example, the isatap "automatic deprecation" and teredo
> "automatic sunset" functions seem to have existing precedence in this
> work.
> 
> Sincerely,
> 
> Fred Templin
> ftemplin@iprg.nokia.com