[linux-elitists] Re: [IRR] Re: Sender-ID

Eugen Leitl eugen@leitl.org
Sun Dec 19 13:55:17 PST 2004


From: "David L. Black" <david.l.black@verizon.net>
To: <somelist>
Date: Sat, 18 Dec 2004 14:24:07 -0500
Subject: Re: [IRR] Re: Sender-ID

>>As the possessor of a number of arrows in my back courtesy of my efforts
>>to (successfully) ensure that the mandatory-to-implement authentication
>>mechanism in iSCSI was amenable to open source implementation, I'm well
>>aware of the existence and importance of open source.  What some of the
>>open source patent policy partisans seemed to overlook in the debate over
>>the IETF's intellectual property policy revisions was that open source
>>is not the only stakeholder, and hence is not in a position to dictate
>>patent policy terms to standards organizations like IETF.
> 
>Well, my opinion is that an internet standard that cannot be implemented
>in open source is a pretty useless standard. ;) But I can see your point
>- -- in some sets of protocols, it's not necessary.

I agree with that view, but I think Justin missed something in dismissing
the interests of non-open-source stakeholders.  I definitely agree that
open source is important and that internet standards generally should be
implementable in open source, but that doesn't end the discussion about
what IETF's intellectual property rights policy should be.  The debate
over the IETF's IPR policy involved a number of disagreements about
what is actually necessary to enable open source implementation,
compounded by some open source partisans taking absolutist positions
of the form "the policy has to be done exactly my way, or else open
source implementation will be impossible".

To take an actual example that came up in the debate over the IETF's
patent policy, it was claimed that *if* there is a RAND patent disclosure
that does not promise royalty-free for a standard, *then* that standard
cannot be implemented in open source, period.  In this absolutist form,
the statement is demonstrably false, and what's even more of a surprise
is that open source has implemented standards covered by patents for
which not even a RAND promise was available - the most obvious example
was open source implementations of RSA for SSL prior to the expiration
of the RSA patent.

This is not to say that any licensing arrangement for any patent from
any patent holder is consistent with open source implementation -
that would be even more absolutist and even more wrong.  Rather,
my view (and the view taken by the IETF IPR policy) is that these
situations vary widely, and it's better to have a flexible policy
that allows a working group to figure out the "right thing" (hopefully
working with the patent holder) than to impose a one-size-fits-all
solution (which rarely does fit all).

Thanks,
--David
...
----- End forwarded message -----
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
Url : http://allium.zgp.org/pipermail/linux-elitists/attachments/20041219/d481ef4b/attachment.pgp 


More information about the linux-elitists mailing list