[OpenID board] IPR Policy and Process Proposal

Drummond Reed drummond.reed at cordance.net
Wed Apr 25 06:56:52 UTC 2007

See [=Drummond] inline.



From: board-bounces at openid.net [mailto:board-bounces at openid.net] On Behalf
Of Dick Hardt
Sent: Tuesday, April 24, 2007 9:25 PM
To: board at openid.net
Subject: Re: [OpenID board] IPR Policy and Process Proposal



On 25-Apr-07, at 2:31 AM, Drummond Reed wrote:



RE the IPR Policy, Blanket Reciprocity term, I don't think you give a
license for anything other than the Necessary Claims you contribute. The
issue is that if you sue another Contributor for ANYTHING - even something
totally unrelated to OpenID - you lose the license they have provided to any
Necessary Claims they contributed.

See the suggestion in the Blanket Reciprocity open issue section for a
simple way to fix that.

If I can't sue, I can't enforce, and I effectively have granted a license.
Even with the suggested change, If someone's implementation infringes on a
patent that is NOT required to implement OpenID, then if I sue, then they
can counter sue for infringing on any OpenID related patent. Effectively I
have granted a right to any patent used in their implementation.


[=Drummond] With the suggested change, if you sue a contributor over
infringement of a patent you own that has nothing to do with OpenID, they
can't sue you back for infringement of their patent that reads on OpenID,
i.e., your suing them doesn't invalidate their "patent promise". Only you
suing them over infringement of something you contributed to OpenID under
the patent promise would enable them to countersue you for infringement of
something they contributed. That's why I like the suggested change.


RE the formal process, I agree that essentially it's a very lightweight
standards process. Call it a community process. But if you read Gabe's
proposal closely, the full skeleton is there. Voting (which would take place
directly on the lists), membership (which involves only membership on the
lists), governance (which is this process doc itself) is all covered - it's
all just very lightweight.


Given the IP significance of this, the voting process is too loose. This is
not open source software where there is a right to fork and on-going support
of a core group is needed to move forward. The dynamics around a standard
are quite different, and my experience with voting on the OpenID list has
been very disappointing and inconclusive.

[=Drummond] I can only say that with this lightweight IPR process in place,
voting on the list would become more formal and "tighter" while still being
a lot less heavy than standard SDOs.


Overall, if the choice at this point in time comes down to putting in place
"just enough process" for the OpenID community to have its own lightweight
process vs. being forced to move the specs to a full-blown SDO, I'd prefer
the former.


IMHO: Given the progress of the community to date, I think we would be able
to produce a better spec faster in an existing standards organization. :-)


[=Drummond] On this we need to take the pulse of the board (and the
community) - are we better of going to an SDO or putting our own lightweight
process in place? I've already stated my preference, and Dick his. What do
others think?


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-board/attachments/20070424/b3d94839/attachment-0003.htm>

More information about the board mailing list