There are a bunch, although many of them are likely not in full compliance
with the RFC, which specifies support for certain aspects of ESMTP.
Remember, the entire BITNET network was implemented using BSMTP to get
around the 8X8 address limit of NJE. In its heyday BITNET amounted to
several thousand systems running on a variety of hardware and software
platforms.
i was an email-only user (i.e. not a user w/ administrative knowledge)
when i was using bitnet, so i cannot claim to have any in-depth
knowledge. thanks for the explanation though.
One implementation that is up to date is the one in PMDF, which is descended
from our BITNET handling code. The same code is also in SIMS.
thank you very much for the pointers.
I also doubt if we would have written the specification had we not had an
actual implementation in hand. The BITNET experience showed that turning an
interactive protocol into a batch protocol has some subtle spots to it.
in your opinion, are those subtle spots adequately and comprehensively
covered in the rfc?
Yes they are. Indeed, this was the entire point of writing the RFC -- to
explain how this can be done and what the issues are with doing it.
i ask because if bsmtp encapsulation is to become
widely used (assuming here that it isn't at the moment), it seems to
me that there may have to be a few more implementations.
Frankly, I doubt if it will ever be widely used. But writing the document
seemed to be useful given that there have been a fair number of implementations
in the past and they didn't get it right without a specification.
Ned