Re: Understanding response protocols
2004-10-13 20:59:00
Actually some people are proposing blind acceptance. They want MFT to
be honored automagically, regardless of the content of the reply.
They want MFT to be honored when the replier uses the Reply-to-List
button.
Well, that's just nuts. MFT doesn't specify the address of the list.
For that matter, it doesn't specify the address(es) of "all" or
"author" either. If the author of a reply tells the user agent "reply
to list" or "reply to author" or "reply to all" that's what should
happen. MFT and Reply-To need to alert a recipient to the subject
message author's request, and make it easy for the recipient to honor
that request, without changing the behavior of the user agent when the
recipient explicitly specifies what kind of reply to use.
People seem to be so accustomed to user agents where "reply" and "reply
all" are broken in this way that they have a hard time understanding
that there might be a better way.
The duplicated message problem is broader than that. And no, MFT
doesn't solve even that subset of the problem - at least, not without
creating other problems.
a) have both list addresses appear in To or CC of the original
message
b) recipients use "reply all" to reply to both lists.
It's also the only way for the replier to respond if he wants to
ensure
that all recipients of the subject message receive the reply. The
replier should not assume that honoring MFT will produce the same
result.
Honoring MFT is not intended to produce the same result. If a MUA has
been
upgraded to understand MFT, then it needs to have a separate
Reply-to-List
button.
Nope, MFT and the list address(es) are completely separate. Nothing
assures that the author who sets MFT intends that a "followup" go to
the list when MFT is honored. If MFT is honored the followup goes to
wherever the author who set MFT thinks followups should go, which might
or might not include the list(s).
Yup, it's a problem. But MFT has several problems also. If an author
has multiple MUAs he needs to configure each of them to be aware of
MFT.
Yes. But if he is not fussy, then he can leave it to the mail-expander
to
insert the MFT according to the list policy.
This violates the purpose of MFT. Having lists rewrite MFT is even
worse than having them rewrite Reply-To.
.... Much
simpler to
avoid sending the duplicates in the first place.
It's simpler in some sense. But it makes repliers responsible for
fixing a problem that is on the sender's end.
What problem on the Sender's end do you have in mind? That he might set
MFT to president(_at_)whitehouse(_dot_)gov?
The sender of the message is the one who is expecting the recipient's
MUA to not duplicate messages for him. That's something to avoid if
you can. What you want - as much as you can - is for each party's
software to be responsible for acting in that party's interests. That
way, when the software doesn't work to your satisfaction, you are in a
better position to fix it. Or to use a concrete example - if you want
your message store or MUA to process duplicates in a particular way you
can upgrade or configure your message store or MUA to do that, and
you'll be happy. But if you want everyone else's MUA to eliminate
duplicates for you then you might wait a long time for something to
happen that is completely outside of your control.
It hides the fact that
the replier intended for certain recipients to receive a copy of the
message. It makes it difficult for a replier to know whether the
recipients omitted by MFT will actually receive that message.
If the replier is fussy, then he will use "Reply-to-All" and edit it
the
hard way.
Again, if we want lists to be more effective tools for discussion then
we need to make it easy for people to do the right thing when replying,
not to arrange things so that only those who are "fussy" will consider
it worth the bother.
It is my belief that an MUA that makes it easy to specify who should
get replies will make its user into a more effective communicator, and
those users will start to see a benefit. Once other users realize
this, they will also want MUAs that work that way.
Keith
<Prev in Thread] |
Current Thread |
[Next in Thread>
|
- Re: Understanding response protocols, (continued)
- Re: Understanding response protocols, Philip Hazel
- Re: Understanding response protocols, Keith Moore
- Re: Understanding response protocols, Charles Lindsey
- Re: Understanding response protocols, Keith Moore
- Re: Understanding response protocols, Charles Lindsey
- Re: Understanding response protocols,
Keith Moore <=
- Re: Understanding response protocols, D. J. Bernstein
- Re: Understanding response protocols, Hector Santos
- Re: Understanding response protocols, D. J. Bernstein
- Re: Understanding response protocols, Charles Lindsey
- Re: Understanding response protocols, Bruce Lilly
- Re: Understanding response protocols, Keith Moore
- Re: Understanding response protocols, Bruce Lilly
- Re: Understanding response protocols, Hector Santos
- Re: Understanding response protocols, Keith Moore
- Re: Understanding response protocols, Hector Santos
|
|
|