Re: audio, checksums, and trojan horses

Keld, my own implementation efforts refute the notion that implementors are
ignoring everything but RFC-XXXX. I am actually further along in
implementation of full support for mnemonic than I am in implementing full
RFC-XXXX support. Mnemonic is much easier, of course, and it has not changed
so much over time. I have also implemented about half of the original
RFC-ZZZZ. I have not implemented the new header support, but it is the
newest thing on the agenda. Note, however, that the new header support is
predicated on base64 and quoted-printable encodings, so I have at least
part of the work done anyway.

I certainly view all these development efforts as being parallel. I cannot
speak for other developers, of course, but I feel that the grouping of
things into a single RFC versus multiple RFCs is largely irrelevant from
a development point of view. The use of multiple documents is an aid to the
IETF process and to consistency and proper grouping of topics. That's about
all it is.

                                Ned