“crowdsourcing”

Someone at Code4Lib (I think?) talked about using ‘games’ to get the public to provide metadata.

I noticed that Google Images is now trying that out.

I still remain skeptical that people actually find this fun and will engage in it for that reason. But apparently I’m wrong, said the Code4Lib guy who I’ve forgotten.

I do think that if you can provide a very easy way where users can provide tags/metadata in mere seconds as part of their ordinary ‘workflow’, many will probably do it as a sort of public service.  Rather than a seperate interface, why doesn’t Google Images just allow you to click to ad tags right on the search result page, with AJAXy code to make it take only an instant?  The Google computers could keep track of when the same tag was used by more than one user on their own.  But I guess that might be harder to protect from malicious spammers (who can certainly use more than one account on their own), which is becoming more and more of an issue in any sort of open access ‘crowd sourced’ metadata.

What are different ways we could use this technique with library metadata?

This entry was posted in General. Bookmark the permalink.

3 Responses to “crowdsourcing”

  1. Sean Hannan says:

    It was Stefano. Specifically, he was talking about Freebase’s Typewriter (http://typewriter.freebaseapps.com/) and Genderizer (http://genderizer.freebaseapps.com/).

  2. I just tried the Google link. It seems simply to be looking at co-occurrence (a random pair of indexers coming up with the same term) as a way to assess term suitability. It is a speed test to see if you and your partner can come up with the same term before time runs out. The question that remains unanswered, I think, is whether terms gathered in this fashion produce retrieval output with improved relevance. I didn’t have the feeling that what I was adding to the pile would do that in a meaningful way.

  3. Shannon says:

    This is really exciting to me – I’ve been asking this same question. Image tagging seems like such a useful tool for catalogers dealing with digital materials. I wrote a little proposal for building such a tool:

    http://web.simmons.edu/~astolfi/xmlpaper.html

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s