Alessandro Vesely wrote:
Michael Thomas wrote:
Steve Atkins wrote:
it is extremely difficult to distinguish wanted bulk mail from
unwanted bulk mail. Knowing whether a particular message was
solicited or not is a very useful data point in making that decision.
Yeah, sure. I'm not even sure "solicited" makes a useful distinction
either, since those arguments usually devolve into legalistic claptrap.
"Unsolicited" still makes the difference between spam and legitimate
bulk email. Having some data that might be held as legal evidence
would be a further incentive for legitimate bulk mailers to behave
correctly.
It might make a difference to somebody, but not to end users. There
seems to be three states to end user thinking:
1) I'm not done with this message
2) I'm done with this message
3) I didn't like this message, and don't want more like it
Note that #3 often coincides with spam, etc, but it is not a
1:1 mapping.
If you think of the "junk" button in terms of "I don't like this"
for whatever reason, we might make some progress in teasing out
people's fickle intents.
Doesn't that stem from the overall UI design? For example, YouTube's
text comments sport a "good"/"bad" pair of buttons, beside a "spam"
button, to classify comments. More buttons, less fickleness.
I'd say that given the volume of email lots of people go through, it's near
miraculous that the current delete/junk differentiator works at all. The
problem
isn't with the two buttons, it's that UI designers have put the wrong
label on
the "Junk" button. Users will not be fooled though: the Junk button really
is the "I don't want this (anymore)" button and they all know it. And it's
certainly not their problem that the UI designers maintain their procrustean
stance that their users are the ones who are wrong, stupid, unreliable,
etc, etc.
In short, these designs need to conform to the way that users manifestly
use them, and that includes putting legitimate but unwanted email into the
current "Junk" folder. It's not their fault that the logic behind that
button
isn't figuring out the best way to gag the unwanted mail at its source if
possible, or build filters which are more selective, and based on better
clues (cf DKIM, SPF...).
Assuming that the feedback loop was completed, of course.
Currently, FBL is only available for extra large ESPs. (About a dozen
of them?) Isn't that a limitation?
With the minuscule amount of email traffic I have, I don't think
anyone would even consider applying to get FBLs from my servers. I
assume that's because entering into an agreement takes a finite amount
of human resources. Isn't it possible to automate the subscription
mechanism, so that users of small ESPs can return that feedback to
bulk mailers too?
Sure, unless we built something where a domain holder could relatively
easily assert itself and have some standardized mechanism to "Get me off
this
list/campaign" all of this FBL stuff is likely to based off of
handshakes between
large entities. The latter is not impossible given existing auth
infrastructure, but
it's not by any means free either. So it would be nice to see a variety
of these
bilateral agreements, how they work, what their efficacy is before
rushing off
to solve a general problem.
So to that end, do, say, AOL and EBAY work together? What do they do?
How well does it work? How well does it scale human/automaton-wise?
Is, say, AOL thinking about expanding access to other EBAY-like creatures?
Is there some sort of 80/20 rule on who they would take and who they
wouldn't?
If there is, what logic drives the 80/20 classifier?
If we can get answers to these kinds of questions over a larger set of ESP's
and Etailors/legitimate bulk senders some patterns _may_ emerge and _may_
point toward standardization making the situation better.
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg