Marshall, > ok, so the current process is adequate, we just need to be a little > more careful in following it, right? Mainly yes, but I'm sure processes could be improved, too. The reason why I sent my initial e-mail was to warn authors. And to ask them to check to make sure they're sending the right file. This falls under the category of "be more careful". Similarly, chairs and ADs should be careful about checking AUTH48 results -- clearly, accidents can happen and people should not just assume that there was a good reason for a change. With regards to process improvements, there might be some ideas as well. For instance, automated tools -- send the XML from the ID submission tool directly to the RFC Editor, eliminating author errors and e-mail roundtrip delays. Other ideas played around in this thread include prohibiting any XML change. Or immediate checking after reception at the RFC Editor side that the file is correct. But I'd be a little bit more cautious here. I'd like to give some room for the RFC Editor to organize and order their work the best way they see fit, as long as all the steps that are needed will be taken care of. And as an AD I would rather deal with one question from the RFC Editor about the appropriateness of changes, rather than separately for approval, RFC Editor's own process, and AUTH48 stages. Jari _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf