[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: draft-rfc-editor-rfc2223bis-04.txt

--On Tuesday, 27 May, 2003 14:27 -0700 Bob Braden <braden@ISI.EDU> wrote:

So, if I understand John, the problem at hand stems from our
decision to to use I-Ds as the channel for individual
submissions as well as IETF submissions.  This was expedient
and convenient, but we overlooked the fact that it has IPR
implications on the document before RFC publication.
See Scott's note, and be careful when using the term "IPR", which has been a lot of the problem. We've got an IPR issue with patents, etc. (what the IPR WG now calls "technology rights"). There, Scott believes that we should force the same early disclosure regime on authors of individual submissions that we force on documents being contributed/submitted into the standards process. I don't have a strong opinion on that subject, one way or the other.

The other issue, also "IPR", has to do with copyright, and what should be required to be given up and when. On that subject, it is, IMO, entirely reasonable for the IETF to insist on permanent licenses for archiving, redistribution, and derived works on anything that is contributed to, or arises from, the standards process (even non-standards-track documents). But, for simple publication/posting of a document as an I-D, there is no necessity --at least no necessity rooted in the standards process-- for any license grants at all that extend beyond the expiration date of the I-D. I.e., the argument about needing extended rights has to do with what I would describe as "IETF documents" (a much narrower usage than in Scott's I-Ds), and not with what posting an I-D which is not an "IETF document" needs to require.

Is that right?  Would it fix the situation if we accepted
individual submissions WITHOUT prior publication as an I-D?
Yes, although that isn't the only solution and might not be the best one (especially given that both the IESG and the RFC Editor has a history of misplacing individual submission documents, more different sorts of entities to track is not necessarily a Good Thing). The general idea would be to accept the _requests_ to publish without any extended grant of copyright-related rights, getting a transfer of those rights only on publication or when you agree to publish. Scott and I agree that far, but disagree on how best to handle it, but I have to admit the distinction is relatively fine.

As I understand it, Scott's preference would be that, at the time the RFC Editor and the IESG signed off on a given document, the author be required to post an I-D that contained an "option 1" or "option 2" statement, thereby transferring the appropriate rights.

I would prefer to accomplish the same end by your getting a release earlier in the process that is contingent on RFC publication. It seems to me to make no difference whether that comes as part of an I-D or in a letter, but its effect is that the permanent license to republish and distribute, and any rights to create derived works, etc., would apply only to the RFC and not to any intermediate drafts that might be floating around (no matter how close their text was, or was not, to the RFC itself). If nothing else, doing things that way reinforces the notion that RFCs are our archival series of record. It also makes it more feasible for an author to "bury" intermediate drafts, which is a subject on which I don't think we need to take a position. I.e., if an author wants to go to the trouble to try to do that, I don't see any reason why it should be a requirement of the RFC publication process to prohibit him or her from doing so. I also consider posting an I-D purely for that purpose to be a waste of time, but that may not be significant.

john