Re: iSCSI: SRP status

[Tom Wu <tom@arcot.com>]

David Jablon wrote:
>>I, David Jablon wrote:
>>
>>>[draft-jablon-speke-00.txt] is careful to not state feelings or legal conclusions,
>>>
> 
> At 08:31 PM 3/27/02 -0800, Tom Wu wrote:
> 
>>After reading the draft, I must respectfully disagree.  Although it doesn't state explicit legal conclusions such as "X is covered by patent Y", it makes statements of opinion like "X uses techniques from Y (which is patented)" or "X is fundamentally different from Y", which, if the reader believes these assertions, leads him/her to a corresponding legal opinion.  I would suggest that opinions w.r.t. IP be struck from the I-D.
>>
> 
> Thanks.  I'll review those statements, and look into removing any
> opinionated remarks.  However, I think it's important and helpful to
> present facts about these methods to assist people in performing
> their own analysis.

[...]

>>Are "unsupported opinions" proper/improper in an Internet-Draft, though?
>>
> 
> What I said applies to drafts too.
> If anyone points out specific language of mine that appears to have
> crossed this line, either privately or publicly, I'll definitely look into
> correcting it.

In my opinion, section 4.10 of that I-D, in its entirety, crosses this 
line and sticks out like a sore thumb.  It presents no new facts about 
the techniques being discussed; the math behind SPEKE is already 
disclosed elsewhere in the draft, while the math behind SRP is disclosed 
in RFC2945.  What it does do, is present the author's unsupported 
opinion on what techniques are similar or different to/from other 
techniques.  I ask that it be removed.

Although I understand the need to justify the existence of an I-D that 
discusses patented, non-free technology, I submit that an I-D is an 
inappropriate place to state opinions about IP, even if it is done under 
the guise of "structural analysis".  More to the point, other I-Ds and 
RFCs, even in areas with IP disputes, like VRRP and the idn drafts, seem 
to have steered clear of including anything beyond a simple IPR Notice 
section, as mandated by RFC2026.  This is, of course, based on the 
documents I've managed to dig up so far; I'd be interested in hearing 
the opinion of someone with more experience in IETF procedure.

> 
> -- David
> 


Tom
-- 
Tom Wu
Principal Software Engineer
Arcot Systems
(408) 969-6124