... > Sure, but there's still an issue here, or at least something needing > clarification: Is RFC 4568 trying to refer to a=key-mgmt when it uses > a=keymgt?
I can only expect that was the intent, yes. This should be noted via http://www.rfc-editor.org/errata.html (which isn't ideal, but it's all we have). Might want to ping the authors first to make sure that was their intent. > It seems like RFC 4566 section 10 (Summary of changes since RFC 2327) > should have mentioned this issue, and others where complying with 4566 > would make you (at least in theory) not interoperable with RFC 2327. > All 4566 says about it is: > > The ABNF grammar in Section 9 has been extensively revised and > updated, correcting a number of mistakes and incorporating the RFC > 3266 IPv6 extensions. Known inconsistencies between the > grammar and > the specification text have been resolved. Without going into a case-by-case analysis of those changes, I dunno if there would be much value in highlighting "-" in attribute names; highlighting it might cause some readers of the errata to assume that was the only change, which could make RFC4566 'compliance' worse (if that was thought by some implementor to be the only substantive change to the grammar). -d _______________________________________________ Sip-implementors mailing list [email protected] https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
