Courtney
Wokai has recently found itself looking up another web 2.0 word: crowdsourcing.
Crowsourcing is when a community of independent people collaborates to accomplish a task. The work of many can replace that of one. And in fact, the workproduct is better. It usually is more accurate, completed in a shorter timespan and remains constantly up-to-date.
James Surowiecki flushes out the philosophy behind crowdsourcing in The Wisdom of Crowds. James looks at why large groups of people are smarter than an elite few, no matter how brilliant. He provides countless examples of why the masses are better at solving problems, fostering innovation, coming to wise decisions, and even predicting the future. Take a task like figuring out how many gumballs are in a gumball machine. Who would come closer: a group of 100 people collectively deciding or one expert gumball guesser making a wager? James bets on the crowd.
And it seems as though netizens'do too. When was the last time you went to the library to find information? Chances are your like me and have not visited a library for over ten years. Instead, you hop online and use world’s most successful crowdsourcing champion Wikipedia. Through the collective genius of many, Wikipedia has succeeded at the task of being to answer just about any question.
As much as I love Wikipedia, I still question the crowds ability to come to the right answer.
I did an experiment. On Facestat, you can receive feedback on any photo of yourself you upload. Here is how it works: upload a picture, choose five questions you want answered, sit back and wait while your photo is judged. In less than 24 hours, you obtain the averaged opinions of everyone that judges your photo. The results are fascinating…
I asked people how old I looked. And the results varied widely from photo to photo. In photo 1, they thought I was 35-39 and in photo 2 they thought I was 22.
Photo 1: I am only 24. God only knows how I will look when I actually am 40!
Photo 2: Apparently I look younger here. My average age being 22. But it crazy that the most popular choices are 18-21 and 30-34.
I think the experiment raises an important question about crowdsourcing: Who is in the crowd? And does it matter? In regards to Facestat, you can see the judges of people who looked at your picture. In the first photo, a bunch of teenage girls judged me. And in the second photo, it was a much older crowd, split half men and half women. Clearly crowd composition played into the results. Maybe Facestat needs a rating and classification system to weigh opinions different. Those who are experts and not? Or perhaps experienced at getting the right answer?
While crowd composition has few ramifications on Facestat (other than hurting users’ feelings), sites like Dermundoand Elanexcreating crowdsourcing applications for online content translation may face bigger problems. With DerMundo, any netizen can translate any webpage. You just click on the pencil in the top right hand corner of a site, up pops a translation box, do your business and submit, and then voila….hours later your translation is live.
Well maybe notquite voila. There are some other complicated back end technical things going on as well. For instance, an automatic translator is often involved. The auto translator can do a portion of the work and the rest can be left up to the netizens. What’s great is that these companies are constantly aggregating all their translations to improve the quality of output from that automatic translator. The more the software is used, the better it gets.
The implications of this are HUGE for Wokai. Anyone anywhere could tap into any webpage and do some translation free of charge. This means that any volunteer whether they be in China or in the US could log onto Wokai, translate a borrower profile, and within hours that translation is live. This would reduce Wokai operational costs over the long term. Great news for a starving non-profit.
But are the implications HUGE for the world? For interest driven sites, I think yes. For product driven sites, I think not. I think that crowdsourcing translation will work for websites build around interested, passionate, and involved user communities. There are a lot of bi-lingual netizens who care about Wokai succeeding, so I have no doubt they will log in and translate. But what about sites like Amazon, Microsoft, Yahoo, etc. Will they obtain dedicated user communities to translate?
I don’t think so. But then again, maybe they don’t have to. Maybe Elanex and Der Mundo can improve the efficiency of their auto translator through interest driven sites and then sell that translation service to the product sites? After all, they are the ones with all the ren min bi and lord knows we are not.
But I guess thats besides the point. What is the point is that crowdsourcing is coming. Its coming in a big way. And it coming directly affects Wokai. We are working on a partnership with Elanex for their software. That is, unless Google does something first..
Recent Comments