> 9 jul 2015 kl. 17:19 skrev Keith Moore <moore@xxxxxxxxxxxxxxxxxxxx>: > > RFCs should reference URNs whenever possible. If there's no URN for a reference but there is a DOI, I have no objection to the DOI being included. If both exist, I would consider it acceptable to reference both. > > I fully support making it easy for RFC readers to access referenced publications. But the RFC editor should not undermine IETF standards efforts. Are you seriously claiming that the RSE is _undermining_ the IETF standards effort by adding an external reference to published RFCs or did I miss something (not being a native English speaker)? > > Keith > > Sent from my iPhone > >> On Jul 9, 2015, at 11:04 AM, Joel M. Halpern <jmh@xxxxxxxxxxxxxxx> wrote: >> >> How about, if we have something better to include, we add that. Then, if we find that the world is using what we produced, and not using DOIs, we can stop using the DOIs. While some folks have commented on list that they do not need DOIs for academic purposes, other folks have been very clear that it is very helpful. >> >> Yours, >> Joel >> >>> On 7/9/15 10:43 AM, Keith Moore wrote: >>> On 07/07/2015 06:00 AM, John Levine wrote: >>>>> Ugh. That seems painful to me. Assigning DOI's to RFCs could happen >>>>> external to the RFC itself, and could be something that doesn't cause a >>>>> lot of friction. But if we are polluting RFCs with these DOI numbers, >>>>> and that we are forced to make an effort to do so as part of the >>>>> agreement with the DOI vendor, I think this whole thing sounds like a >>>>> bad idea and contrary to what the IETF usually stand for. >>>> Here's a thought experiment. What sort of bad things would happen if >>>> the RFC publisher were to add DOIs in the manner described in this >>>> draft for a couple of months? >>> >>> >>> Among other things, it creates the impression that IETF doesn't use its >>> own standards for identifiers. >>> >>> IMO, DOIs need to cease appearing in RFCs. >>> >>> Keith >>> >>> >>> >