On 12/16/09 4:00 PM, Michael Thomas wrote:
I've thought much the same thing for years. To throw something else
out, I think that this is tied up with what I call "the alerting
problem". That is, we are fed many different streams of information
these days (email, im, voice, news...) each of which has their own
filtering and prioritization -- if you're lucky. By alerting, I mean
the means that you become aware (or not) of a piece of communication
vying for your attention.
What I really want is to have the streams classified as is relevant
to me, where the alerting mechanism is tied to its *importance* to
me, rather than the communication channel. That is, there is nothing
inherent with voice that should give it preference to wake me from
my sleep to hear an ad for the latest in garbage pail liners. I want
something to make sense of all of these things and raises things to
my conscious as makes sense for *me*.
Give me that, Nigerian scams and penis enhancement all pretty much
take care of themselves -- my classifier already knows that I could
give a flying-f about those.
Completely agree, when describing feedback originating from individual
users. Your provider might be able to make sense of what is unwanted in
your view by observing messages marked by you as "junk". They might
even extend this service by unsubscribing from mailing-lists, and
tracking repeat violations of expressed desires.
However, not all feedback would fall into the user category. Feedback
obtained from spam-traps that recognize auto-responses and valid DSNs
offer better metrics when separated from user feedback of lesser quality.
The general concept was to offer a conduit service as a means to protect
the provider's feedback address from abuse, while at the same time,
allowing the publication of both user and spam-trap metrics as a means
to generally inform their recipients and management.
-Doug
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg