>>Maybe a poll on user expectations, main version of the product used, or something that isn'tr obvious...
>
>Perhaps all they are interested in is how many people click the X without voting.

Isn't a tradition to use psych undergrads for those types of surveys?