At 08:22 01/02/2004, Dave Crocker wrote:
State the issues in entirely non-technical terms. What enhancements
do we seek, in terms that are visible and important to end-users and
to operators? No reference to protocols or formats is permitted.
Talk only in terms of user interactions and operator actions.
For example, we want users to be able to exchange email in their
native language.
Yes
Actually, AIUI, this is possible now... (Given the amount of spam email I
get in Russian & Chinese, it looks like it..)
Maybe the requirement would be to have more interoperable use of email with
different native languages.
We want them to be able to have addresses that are
expressed in their native character sets.
I have difficulty with this one. OK, I'm English, so may be biased. But if
I received an email in English from someone with a Chinese email address,
I'd have great difficulty handling that email address.. I believe I'd have
more trouble dealing with that email address than the person in China would
have with dealing with my email address. Computers in general don't handle
non-US character sets very well.
So, I can see the use of native character set email addresses within a
locality, but not across character set boundaries - that would cause big
problems. The issue is how to deal with it.
Perhaps we want senders to be accountable, so that problems with their
posting can be traced back to them. Or at least, we want _someone_ to
be accountable for the postings. Or perhaps we want to provide a
means that takes note of senders who are willing to be accountable,
and correctly distinguishes them from senders who are not willing.
Yes, any or all of those.
We want people to be able to exchange email with any type of content.
(Hmmm. They already can do that. So is the new requirement that we
want them to be able to do this with smaller messages?)
That's not one of my key requirements. If a new email protocol is going to
be designed, then it'd be sensible to add this as well, but it's not a good
enough reason to have a redesign.
I'd also have:
- be able to decide whether to accept an email before I get it (ie be able
to examine message characteristics before receiving it)
- Increased reliability, and the ability to trace a message from the
sender's end - ie "has this message arrived?"
A feature of SMTP which is vital to keep, IMHO, is the ease of
testing/debugging the protocol. Most people who've had to debug problems
with protocols like POP3/SMTP/IMAP4/NNTP and also with ones like LDAP will
be able to tell you that the text based ones are easier to handle.
Paul VPOP3 - Internet Email Server/Gateway
support(_at_)pscs(_dot_)co(_dot_)uk http://www.pscs.co.uk/