Re: [xml2rfc] RFC Bibtex format doi numbering incorrect

[John C Klensin <[e-mail]>]

--On Monday, May 29, 2017 3:07 PM -0700 Henrik wrote:

> And no matter how opaque the DOI series identifiers are in
> general, there's still nothing that prevents the RFC Editor
> from _having_ a defined mapping from RFC numbers to DOI and
> vice versa, if a decision is made that this would be helpful,
> right?

Sure.  But it may actually not be a good idea.  Most of the
reasons were discussed at length when DOIs were first being
assigned to RFCs but there is an additional complication now.
As we move into a somewhat ambiguous state about authoritative
versions of RFCs, and different representations contain slightly
different information, or at least organizations of the
information, it is not at all clear that it is desirable for the
same DOI to be assigned to and used with all of the different
forms.  I'm not suggesting that whatever is being done should
necessarily be changed today but, as things evolve, I'd rather
see RFC Editor-assigned DOIs treated as if they were opaque,
giving the RFC Editor the option of changing whatever rules they
use going forward (including potentially assigning different
DOIs to different presentations of an RFC), rather than freezing
a specific mapping or set of rules into an algorithm that is
then embedded in other systems.

If something algorithmically bound to "an RFC", rather than
supposedly to a specific "digital object" is needed, consider
using urn:ietf:rfc:NNNN.   The two actually can, and probably
do, have different semantics and constraints.

I will now return to quietly lurking.

   john