--On Thursday, July 02, 2015 21:01 +0200 Stephane Bortzmeyer <bortzmeyer@xxxxxx> wrote: > On Thu, Jul 02, 2015 at 11:08:12AM -0400, > John C Klensin <john-ietf@xxxxxxx> wrote > a message of 35 lines which said: > >> Maybe the choice of DOIs is more or less arbitrary or >> reflects a too-narrow community of discussion > > There is a very aggressive marketing from the private comopany > which sells DOIs. Like most marketing, it relies partly on > pure lies (for instance FUD against the properties of the > URIs). It seems to be often successful. Just in the interest of separating one sort of FUD from another. DOIs (as distinct from Handle, Handle systems, and several other things) are allocated at the top level by the International DOI Foundation, acting as a Registration Authority under ISO 26324:2012. The DOI Foundation accredits "registration agencies" who. at least in theory, compete with each other (and the I-D calls out the one of those that is being used by the IAB/ RFC Editor). Yes, there has been some aggressive marketing. I haven't seen the FUD about URIs to which you refer. Although I have no doubt that it exists if you say it does, I've generally found the senior people of the International DOI Foundation fairly open to working with others to make different types of ideas and identifiers more compatible (certainly no worse about that than the IETF often is). However, there are enough organizations involved in the DOI world that talking about, or making claims about, "the private comopany which sells DOIs" is almost certainly misleading or at least paints an unspecified entity with two broad a brush. Finally, to address another FUD-like comment (I can't remember by whom), while ISO 26324:2012 is, like most other ISO Standards, associated with a restrictive copyright and a very high sales/distribution charge, as far as I know, all of the significant DOI documentation is available in the _DOI Handbook_, which is available without charge at http://www.doi.org/hb.html. I think there may be reasons why DOIs are not an optimal choice. The decision to use them was apparently made months ago and, while I think it may be appropriate for the community to review how that decision was made and who is accountable and for what, I think it is much too late to say "don't do DOIs". I also think there are some other identifiers that might be equally or more appropriate. People who favor them should probably have a discussion with the RSE (and, as needed, the IAB), about whether it would be appropriate to support their preferred identifiers too. If the answer is "yes", start writing. If it is "no", then, well, that would make a much more interesting discussion for this list, at least for me. john