1) Patents happen, get over it.
Under the current patent system a company that does not apply for patents risks
finding that a patent troll has applied for their idea. This is perjury, it
should be prosecuted. But it is not. And fighting that type of lawsuit has cost
comapnies involved in IETF $5 million or more to have a completely ridiculous
claim thrown out.
So nobody should be appologizing for applying for patents. It is simply a
necessary concern for any major company.
2) Very few patents are so essential that they are worth more than
In the communications world the value lies in the ability to connect. There are
very few technologies that are so compelling that the value of the enhancement
to communication that they bring is greater than the cost of a smaller
communications network. The only example I can think of in the history of the
Internet is public key cryptography and the original Diffie Hellman/RSA
3) It follows that the only allowable patents are on non-essential aspects
If someone wants to use a proprietary encryption cipher with SMTP, let them.
And make sure that the standard specifies how to do so in order that
unencumbered implementations can take advantage of the feature when it becomes
4) In rare instances a standard can turn a worthless patent into a valuable one.
But only if the exercise of the patent is made essential to the communications
role. As in the audio and video codecs that became essential due to being
required for DVD.
Sometimes groups try to spring a patent by surprise. That is irritating. But it
happens much less now that patent applications are no longer secret in the US.
There is definitely a problem with IETF rules in this area. In my view the IPR
regime is a part of the requirements for the spec.
5) Blanket rules give purported patent claims too much power
Most US patents are completely worthless as far as enforcement goes. The main
use of patents is to persuade Venture Capital to part with funds and some small
time trolls extort license fees.
But if the IETF were to adopt a hard rule that said that it would adopt no
standard with a proprietary claim then people will quickly start claiming a
proprietary interest in technologies that they don't want to happen for
There is no practical means for the IETF to adjuicate on such claims. Its easy
enough to propose forming a board, but who would sit on it? Who would have the
legal and technical skills and want to accept the liability?
The IPR working group turned into a farce because the two principal sides that
showed up were (1) the companies with big patent portfolios that they would
rather not have to think about, still less search for potential claims and (2)
professional expert witnesses in patent cases.
Ob. Disclosure: Yes, I do offer consulting and expert witness services in IPR
disputes. This started as a dodge I learned from Alan Shiffman, once you
announce you are offering services as an expert witness in an area, you are
less likely to be dragged into somebody else's patent dispute as has happend
from time to time, or at least if you do get dragged in you at least get paid.
From: ietf-bounces(_at_)ietf(_dot_)org on behalf of Richard M Stallman
Sent: Sun 3/8/2009 6:12 PM
To: John C Klensin
Cc: Jorge(_dot_)Contreras(_at_)wilmerhale(_dot_)com; ietf(_at_)ietf(_dot_)org
Subject: Re: Consensus Call for draft-housley-tls-authz
But an experimental RFC is not a Proposed Standard, a proposed
standard, a document that is in the process of being considered
for standardization, or any other sort of standard or
There are people who propose this as a standard; in factual terms,
that makes it a proposed standard. Whether or not the fact of
publication as an experimental RFC would make it one, it is one
If publication as an experimental RFC entails any sort of approval,
such approval is what a patented proposed software standard should not
Ietf mailing list
Ietf mailing list