Wikipedia, WordPress, Linux…the list of community-driven projects goes on with more and more people getting involved. They do so, it seems, purely for the sake of contributing, bettering themselves and the projects with which they become involved, almost always for no financial reward. This is not crowdsourcing. This is open-source production, collaborative learning, or online community.

Crowdsourcing is using lots of people across the internet, who almost certainly don’t have any direct community connection, to carry out tasks that cannot be done efficiently or automatically using computers…for almost zero cost. Cloud labor you might call it, because the individuals use distributed tools that are held in remote servers “in the cloud” to do their job. A good example of cloud labor is the “game” Google Image Labeler, that millions of people play they label (or tag) random images to help image search engines create an annotated and so more useful database.

Indeed it is possible you have already had contact with Crowdsourcing without realizing since many websites use cloud labor to moderate (i.e. check) the content of photographs uploaded by users (CrowdSifter).

Jonathan Corney, Carmen Torres-Sánchez and Prasanna Jagadeesan of the Department of Design, Manufacture and Engineering Management, at the University of Strathclyde, in Glasgow, UK, and William Regli of the Department of Computing Science, at Drexel University, in Philadelphia, USA, emphasize that the popular conception of crowdsourcing is aligned with the community type projects, but by strict definition crowdsourcing is anything but communal. Cloud laborers do not belong to a group, a corporation or a network and do not necessarily even communicate among themselves, they explain. “Thanks to internet technology, they can access tasks, execute them, upload the results and receive various forms of payment using any web browser,” the team adds. Indeed, “Cloud computing has made irrelevant both the physical location of those workers and where the resources are hosted.”

The term crowdsourcing as defined by Jeff Howe in 2006 is “the act of a company or institution taking a function once performed by employees and outsourcing it to an undefined (and generally large) network of people in the form of an open call”. Again, there is no “coming together” or “self-organization” leading to meaningful participation. If you’re a cloud laborer and you’re not getting some kind of remuneration or solid reward, then essentially you’re a cloud slave.

“Commercial crowdsourced initiatives are clearly for-profit, top-bottom initiatives (i.e. open calls) where a single company not only owns but also sells the results that the crowd (i.e. the large network) generates,” Corney and colleagues say.

There are three main types of cloud labor. The first, are creation tasks. These are probably the most lucrative in which an organization persuades recruits to produce something, a design, an image, essays, whatever, which it then sells on to third parties at a profit. The second category is evaluation tasks, which includes market surveys and feedback calls that generate useful demographic and related information about product use and services. The third category may the more familiar: organisation tasks in which cloud labor is used to configure and annotate data. Examples include recaptcha and Google’s imaging tagging game mentioned earlier, as well website ranking sites, such as StumbleUpon, Reddit, and Delicious, and the excellent Galaxy Zoo, which uses cloud labor to categorize astronomical objects.

There are also three types of cloud laborer: The “any” task individual, which includes the likes of image labeling. The “most people” tasks where readability, functionality, and the ability to rate something are critical, e.g. Stumbleupon web page ranking as well as the likes of iStockPhoto and Crowdspring, which require what might be described as common skills. And, finally, the “expert” tasks, which includes the likes of FoldIt a protein research tool, Wikicrimes, and Innocentive.com (which I first wrote about on ChemWeb back in 2002).

For these three types of tasks and three types of cloud laborer there are, you guessed it, three types of “remuneration”: no money, all satisfaction (which includes all kinds of click workers doing menial “any” individual task voluntarily. Piecemeal work, where workers are paid a flat rate for a set amount of work. Third type is rewarded contribution with a bonus or prize. Amazon’s MTurk (MechanicalTurk) offers cash or gift vouchers, as does Task.cn in China, the freelance portals Donanza.com in Israel and Humangrid.de in Germany and txteagle which pays in airtime.

Importantly, from the commercial perspective and the cloud laborer’s point of view, the nature of the payment offered for a task, no matter how small, influences the status of the intellectual property generated. Companies recruiting cloud labor will usually cover this point so that the laborer’s have no claims on their output, so that the company is free to exploit the IP as it sees fit. This is a significant line of demarcation between true cloud labor and the open source and community activities like Linux, WordPress, and Wikipedia.

What is becoming increasingly apparent is that cloud labor is on the increase, there are literally hundreds of cloud tasks out there ranging from the trivial and uber-geeky to the outstandingly innovative and potentially useful to humanity (most are somewhere in the middle). According to Corney and colleagues, “Several studies have demonstrated that there is a large, responsive work force available 24/7 capable of carrying out tasks of a range of complexity successfully.”

Are you a cloud laborer? If so do you feel like a commercial slave or an addict to the task or are you generously rewarded for your toil? Let us know.

Jonathan R. Corney, Carmen Torres-Sánchez, A. Prasanna Jagadeesan, & William C. Regli (2009). Outsourcing labour to the cloud Int. J. Innovation and Sustainable Development, 4 (4), 294-313

http://recaptcha.net/ – Stop spam, read books

http://images.google.com/imagelabeler/ – A “fun” way to help Google label images and help improve the quality of Google’s image search results

http://www.galaxyzoo.org/ – Users participate in a large-scale project of galaxy research. This project is one of the exceptions to the rule in that it’s cloud labor, but also has a strong community.

Camclickr.org – http://watch.birds.cornell.edu/nestcams/clicker/clicker/index Cornell lab’s ornithological answer to GalaxyZoo

http://fold.it/portal/ Solve puzzles for science

https://www.mturk.com/mturk/welcome Amazon’s MTurk

http://www.donanza.com/ – Search engine for freelance jobs

http://crowdflower.com/ – Labor in the cloud (DoloresLab color categorisation project now redirects to this system)