On 2012-05-05 04:48, Yoav Nir wrote: ... > an obvious idea would be to change the requirements for a new work item from "rough consensus" to "a bunch of people willing to do the work and at least one willing to implement". Some working groups already work like this, but it's not universal. There's nothing to stop a group of people developing a specification as an I-D and prototyping it. They don't need a WG or a BOF or a sponsoring AD. The barrier for spending collective resources (WG time, AD time, RFC Editor time, IANA time) on it should be real, IMHO. On 2012-05-06 04:52, Hannes Tschofenig wrote: ... > My point is that you will not find interest from young engineers to work on > 10 year old topics. You can try it yourself: give a talk at a university and > see the reaction from the students. Pick a lower-layer topic and a topic > from the application layer (some Web stuff). It's true. But the fact is that as in any major technical system, neglect of the infrastructure is a very bad idea. Just consider what happens to a city if it ignores the sewers and water pipes. Sorry to say that the IETF (and the operators who read RFCs) are in the same position as municipal utilities. It's hard to get students interested in sanitary engineering. Brian