01 March 2013

Crowdsourcing for Science

Welcome back. Have you heard about crowdsourcing being applied to science? Getting the public involved, actually participating in scientific investigations, is phenomenal!

I learned about the topic while reading the print version of American Scientist in my customary prone position on the couch, but the article also appears in the online edition of the magazine (see my P.S.).

Crowdsourcing


Crowdsourcing? No, just the start
of a flash mob. (photo courtesy of
Rachel,
www.RachelPhilipson.com)
To be sure you’re with me, I’ll start with crowdsourcing, the process of outsourcing tasks to an undefined, distributed group of people. The term is attributed to a 2006 Wired magazine article by Jeff Howe, but the process is far from recent.

An early example of crowdsourcing mentioned in the American Scientist article is the Audubon Society’s first Christmas Bird Count, conducted in the year 1900. Instead of shooting the birds in a Christmas Side Hunt that year, 27 birders held 25 counts in locations from Ontario, Canada to California, tallying some 90 species. Wikipedia notes that, even earlier, the Oxford English Dictionary had an open call for contributions and received more than 6 million submissions over 70 years.

Nowadays, crowdsourcing need not be conducted online, though that’s clearly the most powerful, efficient and far reaching approach. And that’s the approach that’s being tapped to enlist the public’s help in addressing specific science project goals.


Science Projects

I’m amazed to learn the extent to which people are willing to contribute time and computing power. In one online science project, Galaxy Zoo, participants were invited to classify Hubble Space Telescope images of galaxies by their shape, and that information was used in characterizing the galaxies’ histories. Only one day after Galaxy Zoo was launched, the site was receiving some 70,000 classifications an hour!

What’s your interest? Space? Climate? Lives of the ancient Greeks? How about whale communication? Oh, I know. You’ve read my blog post on bats and would like to help characterize bat calls. Or maybe you’d be more interested in classifying animals in the millions of images collected by camera traps at Tanzania’s Serengeti National Park. You can find science projects inviting your help on these and other topics at the Internet home, Zooniverse.

Wrap Up

This all reminds me of when I was teaching. Enrollment in my classes was upperclass and graduate students from various disciplines. Student answers to my assignments or quizzes were usually right, wrong, right and wrong or somewhere in between. Sometimes, however, student answers were filled with new interpretations, knowledge or insight, and I became the student.

I thanked and gave extra credit to the student who enlightened me and shared the information with the class. Class discussion of those answers would build on that new base. It was a mini, non-Internet version of crowdsourcing, so-called Wisdom of the Crowd, and it worked.

Thanks for stopping by.

P.S.

Anon 2012. Interface facts. American Scientist. 100:6:463-465:
http://online.qmags.com/AMS17717547?sessionID=627D3AAFBEE914D0273A6D18B&cid=2241017&eid=17547#pg33&mode2
Jeff Howe. 2006. The rise of crowdsourcing. Wired. 14.06: http://www.wired.com/wired/archive/14.06/crowds.html
Wikipedia write up on crowdsourcing:
http://en.wikipedia.org/wiki/Crowdsourcing
Internet home of science projects, Zooniverse:
https://www.zooniverse.org/

1 comment:

Bill Teng said...

Last time I personally got involved in one of these crowdsourcing events was the search for Jim Gray, who was lost at sea on January 28, 2007 and never found. A request came out for experienced image interpreters to analyze imagery from many sources, to try and locate Gray's presumably capsized boat. We used Amazon Mechanical Turk, https://www.mturk.com/mturk/welcome.