On Tue, Feb 02, 2016 at 08:19:16PM -0800, Timothy B. Terriberry wrote:
Ron wrote:
And we are specifically not proposing to give *anyone*, open source or
otherwise, "the option to take an RFC and “run with it”" ...
We're trying to find consensus on the best way to apply BCP 78 and the
options outlined in RFC 5377 to the needs of this document, in a way
that best matches the needs of its intended users to the best interests
of the IETF.
It seems to me like putting a BSD copyright header in an XML comment at the
top of the XML source file and dropping Section 13 from the draft would
resolve all of the issues, if everyone agrees that is an acceptable thing to
do. Ron?
It leaves us with the issue of creating a perverse incentive for some
parties to prefer distributing the "non-rfc" version over the official
RFC - even when they themselves have not modified it and don't have
any direct plans to. And in some cases, will force their hand, making
that the only version they can redistribute ...
But aside from that, I think it would cover the practical problems
which we currently see for implementors needing to use this in their
own work. Which is the most immediate issue here.
It's not my first preference, and not what I think serves the IETF's
interest best, for all of the reasons I've outlined earlier in this
thread -- but if we can't get consensus for adding a RFC 5377 4.3
exception to add the "Rights Granted for Implementing" to it, and
nobody has a better suggestion, then I can accept this option as the
pragmatic way we should proceed here.
I think both options are in the scope of current recommendations to
allow, but I'll defer to peer review on that.