--On 17 December 2009 12:33:59 -0500 Nathaniel Borenstein
<nsb(_at_)guppylake(_dot_)com> wrote:
On Dec 17, 2009, at 11:27 AM, Ian Eiloart wrote:
Twitter seems to think that users are smart enough to distinguish
between "unwanted" and "spam". They give you a button for each. It's an
important distinction that most people can make.
Twitter isn't always right, and my intuition differs from yours on this
one. Fortunately it's something that could be resolved empirically. I'd
like to see such a study, because it wouldn't take very many users who
*can't* properly make that distinction to render the two-button solution
counterproductive. I'd rather have one bit of meaningful data than two
bits of muddled data. -- Nathaniel
Well, then you have to make a judgment about what to do with the data.
Application of fuzzy logic is essential here. You don't ban everyone who
get's hit with a spam report, you wait til the reports hit a certain
threshold, or you feed the message into your Bayesian analysis engine, or
something.
What you can do, though, is stop delivering mail from the reported source
to the recipient, perhaps after asking for confirmation. You need to be
careful that this can be undone, though.
Yes, users make mistakes, and they have differing judgments about what is
and is not spam, and those judgments may not be your judgments. But, that
doesn't make the data useless, it just means that you have to think
carefully about the application of the data in order to keep your false
positive rates down.
Anyway, the worst case scenario is that users hit one of the two buttons at
random. In that situation, you simply treat the buttons as equivalent, and
take the same action for each.
--
Ian Eiloart
IT Services, University of Sussex
01273-873148 x3148
For new support requests, see http://www.sussex.ac.uk/its/help/
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg