3. Blind acceptance of MFT when composing replies is a bad idea,
for the same reasons that blind acceptance of Reply-To is a bad idea.
Well, I'm sorry, but 99% of the time when I reply to emails I blindly
accept Reply-To. I don't have the time/patience/concern/whatever to
scrutinize the recipients of every reply that I send.
What I'm wanting is user interfaces that make it easy for repliers to
put the same degree of time/patience/concern/whatever into the
recipient list of replies as they put into the content of replies.
Right now our user interfaces encourage repliers to _not_ pay attention
to the recipient lists.
I am reasonably certain that such software would need to have the
option to do "blind replies" for people who cannot be bothered to read
the recipient lists. But it would be nice if MUAs didn't presume that
everyone wanted to be so careless.
I assume that the software will "do the right thing". And I suspect
that I am pretty typical in this.
Software cannot "do the right thing" because the software has no idea
_who_ you are replying to and whether that is the appropriate set of
recipients for the content in your reply.
What if there's no Reply-To? Is "blind acceptance" of From a bad idea?
If MUAs hid From from you when you read a message, it would be. But
usually From is prominently displayed both in the message itself and in
the summary view, so it's rarely a surprise when replies go to the From
address. (It can still be a surprise if the MUA shows only the _name_
portion of the From field, and the address isn't the one that the human
reader normally associates with the name).
Keith