Thanks Bryan, > The good news is we have a version control system that > will let us track exactly when and by whom the change was > made. And we've all been quite good at adding comments > to our commits. Oh I now where/when the change occurred (rev 310): "normative language rehaul done detailed resegmentation requirements done some more consistency fixes and editorial actions as per comments" Since it's not a language rehaul, it's not about resegmentation, and it's certainly not editorial, it has to be a part of the "some more consistency fixes" done in that revision. I'm not blaming the editor: he's just an instrument of the overworked TC members who don't have the time to truly check everything. I'm just wondering how many other fixes like that have been introduced since csprd01. My main point here is to underline the importance of having, at the very least, one implementation for every single feature we have in the specification, or we will miss issues such as this one, or issues like the non-implementable preservation of fs/subFs in <cp>. And those implementations need to be done before 2.0 is set in stone. Cheers, -yves