On Oct 10, at 04:34 PM, Eric S. Raymond wrote:
D J Hawkey Jr <hawkeyd(_at_)visi(_dot_)com>:
But I think I can safely say that fetchmail started as, indeed, has even
grown moreso into, more than just a "procotol gateway".
If that's true, it was a serious error on my part for which I am now
paying with pain. I don't intend to compound it.
:-)
Perhaps it's my interpretation of "procotol gateway". When I hear that,
I think of something more akin to an SSH-wrapped protocol than a protocol-
savvy process; SSH doesn't care what it passes after authentication, but
fetchmail does.
Granted, I don't think fetchmail can take the "transparent wrapper"
approach that SSH can.
But it does seem to me it could step aside [more], let the local MTA/MDA
do it's thing with whatever comes its way, and pass back what the MTA/MDA
returns [more] transparently. Having said this, I do know how feeping
creaturism happens; I also turn down ideas and patches, or implement them
hesitatingly (I maintain VTWM).
L8r,
Dave
--
______________________ ______________________
\__________________ \ D. J. HAWKEY JR. / __________________/
\________________/\ hawkeyd(_at_)visi(_dot_)com /\________________/
http://www.visi.com/~hawkeyd/