Bob DiPasquale, application developer by day, comedy writer / caption contest entrant by night, was consistently bothered by the selections of caption contest judges and thought there has to be a better way. Why should the inputs of so many be judged by the few, especially when the merit is so interpretive and the status of humor assessment expert is well... a joke? Why not make it so entrants act as judges? Why not send samples of submissions to the submitters themselves and let them sift the best ones to the top? That's crowdsifting, and here's how a typical visit to Bob's HumorQ.com ' s caption contest works. You sign in to let the page know who you are and it shows you your HumorQ score and how many times your recent submissions have been both presented and selected. You choose to play today's Caption Contest, and it first shows you the cartoon from 3 days ago with 5 captions and asks you to choose the one you think will eventually be the most popular of those 5 (if you're right, you get bonus points for that). You do the same for two more cartoons for yesterday and the day before. Now that you've voted, you get to make your own entry for today's caption contest. Overnight the database works moving selected captions to the next days presentation table and scoring how everyone did. The result, Bob says he can put a number between 1 and 200 on how funny each of his members are at http://www.humorq.com, and so he does.
Wait a minute. What just happened? Hypothetically, 1 billion people could submit ideas to this sort of tool, and those 1 billion people could find the best of those ideas in a few days or even sooner. Wow! Brainstorm this. I was taught to brainstorm with yellow sticky sheets and a whiteboard. This seems like the whiteboard just became the internet, and all we need now is some great important questions to ask that can be answered with mass opinion. Bob explains "I think I stumbled on to something that people a lot smarter than me could make great use of."
Though crowdsifting could be used to to find a better "American Idol" singing talent in a day, that probably won't ever be done because it's not commercially attractive. And while finding the best humor and singing talent could have significant value, the real value in crowdsifting probably lies in how government could apply the tool to find the most popular discussion submissions on an issue in a very short amount of time. A government could ask the people for the best 50 word suggestion on how to deal with a crisis, and get the people's answer in a day.
Using brainstorming with crowdsifting, that's powerful stuff.
# # #
A website that uses caption contests and a process called crowdsifting to put a number on how funny people are.