When Search Fails: Humans To The Rescue
A few days back Stowe drew my attention to flickchart.com - a taste-driven movie list generator. I've signed up and await the credentials to give it a try for myself.
I'm interested because I think it overlaps with something closer to home for me: ditto.net
Ditto.net is being built by a brilliant team of people who work for the same company I do. So I'd encourage you to take all of the following with the required-sized pinch of salt. (I've written about ditto previously here, here and here and here).
But I'm going to give you the heads-up on it anyway because I think it's pretty cool. I'm a fan.
The unedited audio you can listen to by 'watching' the video at the end of this post has me asking one of the founder's, Colin Kennedy, about the project. Watch it if only to be amused by a series of still images of Colin (the result of some youtube mangling of the video shot on my N73!) Hell, we're keeping it real! Luckily the listening is the important bit - and that works fine.
Ditto goes a bit further than flickchart - it applies human sorting to huge catalogues including movies, but also including music, games and... in development right now, web video.
It is driven by the belief that finding entertainment ought to be entertaining. It's not another social network, nor a vertical taste-matching network.
It's out to solve one of the big problems of search - that the buckets we heap stuff into currently break.
"Search fails. When you search for 'best Bob Dylan album' the experience is worse than you might expect. We think human sorting can create huge relevancy," says Colin.
Ditto is therefore not about rocket-science-quality algorithms.
"I was more interested in the quirky stuff that only humans come up with. Lists of the '10 movies you shouldn't watch before a meal' variety," says Colin - exactly the kind of cover lines he used to write as a magazine editor.
"I like the experience of browsing through a record shop, or through a friend's dvd collection, we wanted to capture something of that."
Colin sees the recommendations that algorithms tend to throw at us as something like an over-bearing shop assistant - or a crowd of them. Not very welcoming. Instead he would rather:
"We put you in the right part of the store and let you browse."
Gagging to know what you think... and the guys have a blog for you to feedback directly.

Comments