>I think I know what you are thinking about, but I was thinking of a much simpler system with only two (or maybe three) possible answers (Yes, No, Doesn't matter to me).
>
>Example:
>Q: Do you prefer a GUID as a PK over bigint or int?
>A: Check one: Yes , No, Unsure
>Result: 10% Yes 89% No 1% Unsure (2,000 people responded)
>
>Very un-scientific but maybe a little usefull :)
>
>Thanks for your vote.

Only one or multiple questions, it is the same survey infrastructure. We are releasing a survey for Universal Thread Magazine at the beginning of January. This will give you a good idea of the possibilities.