ietf-asrg
[Top] [All Lists]

Re: [Asrg] Generally workable ways of adding a spam button to MUAs

2010-02-08 11:55:00
Ian Eiloart wrote:

--On 5 February 2010 15:20:46 +0000 John Levine <johnl(_at_)taugh(_dot_)com> 
wrote:

Actually, we've seen a reasonable suggestion a few messages back that
would work equally well with POP and IMAP: extract a reporting address
from the message and send it an ARF report.
Which is certainly not at all simpler than setting an IMAP flag on a
message. And (if it's what the user desires) moving it out of the INBOX.
Can I ask people to stop making the newbie mistake of assuming that
all mail systems are, or should be, just like the mail system they
happen to use?

I'm not making that assumption.

By insisting on MDA-specific methods, it forces the same result as making the assumption that all the world is IMAP (+ POP or whatever other protocol we deign to notice). It isn't.

I'm simply claiming that the problem isn't a single problem (1. "how to report junk messages"). It's two problems: "2. how to report junk messages to an IMAP mailstore operator" and "3. how to report junk messages to a POP mailstore operator".

Four: SMTP
Five: Webmail
Six: Microsoft RPC

etc. We use all five. Actually more, but, those are very much in the weeds.

There is no need to make it MDA specific, everything you can do with direct MDA feedback you can do with asynchronous MDA-independent methods.

Someone else seemed to imply that there's an inherent difference between "reporting for complaint handling" and "forwarding for filter tuning", and MDA-specific methods can only handle the latter.

Nothing could be further from the truth. We do both, via non-MDA-based feedback.

While it might be a bit more convenient for some installations to tune _some_ kinds of filters directly with direct MDA submissions, in many if if not most cases it makes no difference because it isn't the MDA doing the filtering, and there's no reason that a non-MDA reporting destination can't adjust the MDA if it actually needs to.

Indeed, tuning MDAs directly (and independently) can introduce severe operational inconsistencies and manageability problems.

If we want acceptance of this, we need to make it as universal and technology independent as possible.

IOW: while MDA-specific mechanisms can sometimes make part of what you're doing a bit simpler, there's nothing that MDA-specific can do that MDA-independent can't, and MDA-independent is vastly more universally applicable. Especially in mixed environments.
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg