I've heard a number of different groups say that they plan to make semantic
distinction based on selector. For example, they intend to send transaction
mail under one selector and marketing mail under another. Their intent is to
have reputation services distinguish between one domain+selector and another.
If you're thinking about the DAC stuff at domain-assurance.org, our
current spec is for certification, not reputation. A critical
difference is that certification is done with the cooperation of the
certify-ee. This means that the certifier can have inside info about
the way the certified domains uses its selectors, so they can say that
they vouch for this set of selectors that this part of the company
uses to send its kind of mail, but not the rest of the company. The
message types are sender assertions that the mail using those
selectors is supposed to be transactional or whatever, so if you see
mail of some other type, someone's lying.
Reputation systems can't make those assumptions, and I agree that
there's a lot less you can do with a selector. I suppose a reputation
system might be able to say "here's our opinion of the mail we've seen
so far with this selector", but I don't see any way to tie selectors
to message types without the sender's cooperation.
NOTE WELL: This list operates according to