"Bonatti, Chris" <BonattiC(_at_)ieca(_dot_)com> writes:
- We want low throughput mobile devices, but we want to be able
to send large files.
- We want strong user authentication and tracing, but we want to
support anonymous traffic.
- We want simplicity, but we want flexibility and extensibility.
- We want backward compatibility, but we want a clean slate.
- We want instant delivery, but we want our messages saved till
we get back.
Highlighting these tensions is a useful exercise - thanks!
I find myself wondering if we aren't scoping requirements for
more than a single application.
My current feeling is that we can meet all these needs in a single
protocol.
The only issue I'm unsure about is backwards compatibility, but I'm
disciplining myself not to think about that until other details are
worked out, as per the "clean slate" philosophy of this list.
In any case, I'm not sure it's as important as you might think. My
non-technical friends leave messages for each other as entries on
weblogs, or as replies to such entries, or via talker protocols, or a
variety of other means none of which gateway into email in either
direction. If a new protocol offered enough advantages, lacking
backwards compatibility wouldn't inevitably destroy it.
--
__ Paul Crowley
\/ o\ sig(_at_)paul(_dot_)ciphergoth(_dot_)org
/\__/ http://www.ciphergoth.org/