I think it might be unreasonable to expect a particularly accurate
measurement of spam characteristics. There are a lot of potential biases in
any sample that is constructed.
However I think we could hope to create a series of measurements that show
long term trends in various characteristics. So while a measurement of the
proportion of spam that is 419 scams is going to be heavily biased by the
means used for measurement a series of measurements taken in the same way
could provide highly accurate trending.
I think it will be very useful to have measurements that allow the
effectiveness of anti-spam measures to be judged as they are implemented.
Phill
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg