see this blog entry from "joel on software"
for a discussion about cataloging a library:
>   http://discuss.joelonsoftware.com/default.asp?joel.3.378798.0

although that topic has much relevance here,
the reason for this message is a comment on
that entry that pointed to this video of a lecture:
>   http://video.google.com/videoplay?docid=-8246463980976635143

the talk -- about "games with a purpose" -- is the
most interesting lecture i've heard in a long time,
and is a rather stunning exposition of some of the
things i've talked about in the past here, in regard to
summoning the power of distributed effort on the net.

one such post was on "the amazing turk" at amazon,
which farms out tasks requiring human intelligence
to people who are paid a small amount to fulfill them.

this lecture goes one step further by turning the tasks
into _games_, which people play just for the fun of it.
the twist is that, when the game is structured correctly,
people furnish very-high-quality data, and _lots_ of it.

given the difficulty of the challenges already taken on
-- one big example is labeling images on the web --
plus the success achieved, something as dirt-simple as
the correction of typos in o.c.r. results is well-assured;
and even much more difficult problems like translations
suddenly seem much more within reach than they were...

but the most fascinating aspect of all of these applications
is that people are creating valuable data in the process of
_playing_games_...  they're not just volunteering their time
out of a sense of duty or obligation.  they are having _fun_.

without impugning the nobility of any altruistic volunteers,
it seems to me that fun is a much more dependable motive.

all in all, his research shows luis von ahn is a sharp cookie;
his lecture is 50+ minutes long, but it's well worth the time.

-bowerbird

>   www.captcha.net
>   www.espgame.org
>   www.peekaboom.org