Very interesting article from Wired. This is something I have struggled with personally. Is it worth investing the time to automate a process, or is it just cheaper to outsource the smarts of up-and-coming countries? Ultimately I’ve found that not only is it cheaper to outsource the work, but the quality of the results is much higher. The main problems I’ve found are scalability and training. Overcome these for your task at hand and the benefits are immense.
The vogue for human curation reflects the growing frustration Net users have with the limits of algorithms. Unhelpful detritus often clutters search results, thanks to online publishers who have learned how to game the system. Users have tired of clicking through to adware-laden splogs posing as legitimate resources. And unless you get your keywords just right, services like Google Alerts spew out either too much relevant content — or not enough.
Again, I have to say, the quality of the work just blows any automated stuff I’ve done out of the water. However you do have to manage your sources – something like Amazon’s Mechanical Turk is a bit hit a miss, whereas something like Elance allows for a feedback system and a more personal relationship – just not the sheer bulk of work.
What the article doesn’t cover is the fact that most of this type of work is outsourced. It would be very interesting to see what the demographics of the workers are like for something like Mechanical Turk. So if this continues to grow in popularity, what are the long term effects of this going to be? Will this help improve the skills of the contributors or just burn them out with mindless work? I personally think the former – most of the projects I have seen are actually very interesting. I know of people who use Mechanical Turk for fun and as a timewaster – certainly not as an income source.
You can take a look at the demographics of Mechanical Turkers at http://behind-the-enemy-lines.blogspot.com/2008/03/mechanical-turk-demographics.html