[OpenISO] Patents.

Norbert Bollow nb at bollow.ch
Fri Sep 7 16:06:12 CEST 2007


Kenneth <kenneth.info at gmail.com> wrote:

> Lets see if this works . .  .

(Administrative note: Since you posted this from a non-subscribed
email address, the posting was at first held for manual approval
by Mailman.  I've approved it and whitelisted the sender email
address.  Unless you're already subscribed with a different email
address, I would encourage you to subscribe to the mailing list.)

> I personally think that requiring a covenant not to enforce patents
> may discourage some organizations from the OpenISO standards process.

Yes, this is certainly true.

What we need to figure out is whether the disadvantages of the
requirement outweigh the advantages or not.

> Patents, if handled in a transparent manner, can be delt with.
> 
> For exmaple, take the MPEG-2 standard.  [..]

I agree that *if* (like in this case) patents are disclosed and
RAND-licensed, they don't pose a problem of principle for vendors
of *tangible goods* which implement the patented ideas.

However for standards about anything that can be implemented in
software for general-purpose computers, I think that the ISO
requirement of tranparency plus RAND licensing is clearly not
open enough, because it does not guarantee the ability to
implement in Free Software.

In many areas of information technology, Free Software is the
main competition to a dominant near-monopoly vendor.  Isn't it
therefore obvious that it is totally unacceptable to accept
any specification as a "standard" which for patent reasons
cannot be implemented in Free Software?

> I propose a ranking system for the intellectual property (IP) issues
> like the ranking system for the technical issues:
> 
> 1 - intellectual property issues unclear and possibly dangerous
> [Example: the OOXML system may earn a 1 because full implementation of
> the standard could potentially infringe MS patents and it is unclear
> what patents MS covenents not to enforce.]
[ Comment about the example: An additional issue with OOXML is that
it is unclear whether the MS "open specification promise" actually
means anything substantial in the OOXML context, because that promise
is only about "requirement elements of covered specifications" and if
you read the OOXML specification, very very little is actually
required there of conforming implementations.  I have corresponded
with Microsoft about this point, but they have been evading the
issue by consistently responding to my questions with true statements
that didn't answer the questions. ]

> [2, 3, 4]
> 5 - intellectual property issues very clear.  [Example: a standard
> whose technical contributors have expressly agreed not to enforce
> patents; a patent pool is established; known, predictable licensing
> fees or free.]

I think that such a ranking system is a very good idea, and I would
encourage suggestions of objective, verifiable criteria upon which
the ranking could be based.

By the way, I think that such a detailed ranking system is only needed
for patent issues.  Is that what you mean?  (I do not think that in
the present context it is helpful to group patent rights with other
rights like copyright, and "intellectual property" is anyway a
propaganda term for a specific ideology regarding rights about
immaterial goods.)

> A standard with a high technical ranking but a low IP ranking would
> encourage the patent owners to contribute to the pool and offer free
> licenses or non-discrimatory licenses at a reasonable royalty to
> anyone - all for the purpose of improving the IP ranking of the
> standard.

Would this mechanism be more effective than the more radical position
of requiring a patent non-assertion covenant which covers all
implementations of the standard?

> You could even establish "defensive patent pools" where a company says
> that they will use their patents offensively against anyone who
> enforces a patent agains the standard.

Yes... while this is not something that I'd personally be inclined to
work on, if someone is interested in working on under the OpenISO.org
umbrella on establishing such defensive patent pools, I'd support
such projects.

Greetings,
Norbert.


-- 
Norbert Bollow <nb at bollow.ch>                      http://Norbert.ch
President of the Swiss Internet User Group SIUG    http://SIUG.ch
Working on establishing a non-corrupt and
truly /open/ international standards organization  http://OpenISO.org


More information about the Discuss mailing list