On 2 Nov 2020, at 11:24, Dave Crocker wrote:
On 10/29/2020 10:49 PM, Pete Resnick wrote:
it reuses existing pieces of infrastructure that seem to fit this
use,
Except that, really, this use is semantically quite different from the
intended use of MDN (RFC 8098):
"...used by a Mail User
Agent (MUA) or electronic mail gateway to report the disposition
of a
message after it has been successfully delivered to a recipient."
The goal of indicating a recipient's reaction is quite different from
indicating disposition of a message.
I don't disagree. However, the semantics provided by MDN is that of
associating a disposition with a particular message, and it has all of
the machinery to do that association. Associating a reaction with the
original messages seems to me much the same as associating an indication
that a message was "read" or "dispatched" or "deleted". Yes, it
definitely extends the purpose of MDNs, but it doesn't seem like a huge
stretch.
In structural terms, the MDN is also a more constrained environment
than a general email message. Hence, using it for reactions limits
the context for including reactions.
How so? You do have a user-readable part to include a message to the
user, which can be any sort of content type.
Adding the ability to have a reply message include content tagged as a
reaction adds to the semantics of email exchange.
When I think of reactions, I think of providing a UI to add a "Like"
button or similar. That seems separate to me from replies, and in some
cases replaces it. Having the the user readable part allows you to do a
reply should you desire, but I suspect the primary kind of
implementation for this kind of feature is a one-hit button for
indicating a reaction to a message, similar to the sort that I've seen
for implementing "I read this." Both MDNs and a reaction strike me as
"add this adornment to the sender's message" rather than "send this
message as a reply to the sender's message". MDNs seems semantically
closer to the purpose, and the implementation of adding a reaction, both
on the sender's side and the receiver's side, seems closer to MDNs.
Having a message designed for signaling information, about message
handling, be modified to include end-user reactions distorts its
original purpose and limits the added function.
As I said, I agree that this is a change in purpose for MDNs, but I
don't see it as particularly horrible distortion.
But as I said at the beginning, I can certainly live with Ned's
proposal. MDNs just struck me as cleaner.
pr
--
Pete Resnick https://www.episteme.net/
All connections to the world are tenuous at best
_______________________________________________
ietf-822 mailing list
ietf-822(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf-822