Hi,
I read the document, I think it is fine, and it removes some busywork.
Firstly, no one reads the boilerplate. The reader is there because they already have to implement the thing, or at least suspect that they might.
If called for, I find that most authors write something like "this is a really rough sketch" in their own voice.
I understand the concern that it's possible to "game the system" here, by just leaving things in I-D form. But this already happens, because the IETF process is so slow. I do not have any ideas on how to speed that part up, so it's not a criticism. It's just a description.
thanks,
Rob
