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

[John C Klensin <klensin@jck.com>]

--On Monday, May 26, 2003 15:08 -0400 Scott  Bradner
<sob@harvard.edu> wrote:

>> By contrast, the rules and
>> requirements developed for the third category should preserve
>> the rights of an author to publish a document elsewhere if it
>> is not accepted as an RFC and, in particular, the "not
>> previously published elsewhere" requirement that is typical of
>> many journals.
> 
> I do not understand this issue
> 
> 1/ the author has that right now and has had that right - this
> is not restricted by the IETF or RFC Editor in any way by
> publishing an  ID or a RFC
> 
> 2/ if a publication currently thinks that an ID is a previous
> publication I doubt that there is anything we can do about it

Scott, one of us is being obtuse.  It may be me, in which case I
apologize.   Once more, in the hope that it will help (then I
give up)...

(1) There is no possible reason, for a non-IETF document (i.e.,
not standards-track, not submitted through or by a WG) for the
author to have to release permanent republication rights to
someone else until those rights have to be yielded to ISOC and
the RFC Editor as part of the publication process for an
_accepted_ document.  Now, the IETF can insist on such releases,
but doing so is, IMO, sheer arrogance -- the RFC Editor has
never insisted on anything resembling such releases except for
documents that are to be published.  This is partially just a
matter of principle, but partially real, because...

(2) Publications (journals, etc.) do ask for transfers of
copyright as a condition of publication, but normally ask only
that the author will guarantee to make the transfer, and has the
rights to do so, before they accept the paper (or whatever).
But they usually insist that the transfer be complete -- i.e.,
that no one, including the author, have republication rights on
the document without their permission.   In order to make that
transfer, the author has to control _all_ rights in the
publication, which means that the ISOC doesn't have permanent,
irrevocable, republication and redistribution rights.

I have fended off publications (indeed, am in the middle of one
such discussion now) about I-Ds being prior publications or all
or part of the content on the grounds that they are _drafts_
circulated to a community of experts for comment, that they
expire, and that I have granted no republication or archival
rights that extend past the expiration date.  They know, and I
know, that there is no "numbered copy" or "disappearing ink"
ritual involved with this -- there may be copies out there for a
long time.  But those copies are identified as drafts and, in
theory, the copyright holder can take action against anyone who
republishes them.

But, if I can't transfer the copyright, without prior
irrevocable licenses to ISOC, it is all over.

To that end, "republish the I-D with a different option" is not
an option either, unless the RFC Editor (and, as appropriate,
the IESG) are willing to consider the document for publication,
and make a firm decision to accept it (or not) based on an
"option 3" document.   Otherwise, you are, again, asking the
author of a non-standards-related document to give up what might
be important rights in return for nothing.  And that just isn't
morally reasonable.

And, while it is more philosophical than specific, similar
issues apply to _anything_ I write that I put out there as a
draft with an expiration date.   As I've said several times, I
think the IETF needs, and should have, stronger rules, and
granting of rights, to protect the standards process for
documents going into that process.  But those rules are really
about documents coming into the IETF, not any document posted as
an I-D.

regards,
    john