Comments for Why Facebook Should Introduce Filtering

Why Facebook Should Introduce Filtering

2007-10-08T22:29:01Z

Tim O'Reilly and team have done some interesting research into the rise of Facebook as an application platform. They released a report entitled The Facebook Application Platform, which we've been looking at in the virtual offices of Read/WriteWeb. One of the findings in the report is that Facebook apps follow the now familiar power law of popularity - a point also made by Alex Iskold in a R/WW post at the end of September. As the O'Reilly report put it: "a Facebook developer needs to build a huge viral hit to get anywhere near the top. As the aggregator, the company that benefits the most from niche applications is Facebook itself."

Top 10 Facebook Apps Usage; chart by Alex Iskold, 27 Sept 2007

]]> As Alex Iskold put it, "the rationale behind the rush to build Facebook applications is that new applications have a potential to spread virally across Facebook's already mammoth user base." However Alex says that there's "a big flaw in this argument", which is that it has brought too much clutter and overwhelming choice into Facebook.

Alex's suggested solution was controversial - it got some kick back in the comments - but in light of the O'Reilly report it is worth considering:

"Could Facebook do a bit more filtering for us? That might make the site more useful. Why can't there be a relevancy algorithm that cleans up the mini feed? Why can't there be fewer applications that do the same thing? Maybe there should be a standard profile view, which would arrange things on people's profiles in the way that I like - that would surely make it easier to comprehend things. And most importantly, maybe Facebook should refocus on what social networks are about - communication."
[emphasis mine]

We are hearing a lot about recommendation and other filtering technologies today, so why can't Facebook use some of those technologies to make it simpler for users to choose suitable web apps? And it needn't be based on popularity either - there can be ratings for quality and usefulness, similar to Netflix and Amazon.

Incidentally, if you are looking for recommendations for Facebook apps, check out Read/WriteWeb's series of posts on the best third party apps on Facebook. R/WW's Josh Catone picked 50 top apps, across 5 categories:

Comment from Deepak on 2007-10-08

2007-10-08T23:06:27Z

But don't we have a built in filter; your contact list? I am much more likely to install applications that my friends install. In fact, that's pretty much the source for any app that I am currently using. While the "bacn" element to apps is a little irritating, the filter provided by your contact list (or social graph if we want to stick to the times) should be sufficient

Comment from Eli on 2007-10-08

2007-10-08T23:21:31Z

That graph that Alex Iskold published is misleading and incorrect. You should stop publishing it. You can't sum daily active users across a number of applications because it does not account for a user that is using multiple Slide apps in a given day. It also does not account for the fact that RockYou has a number of apps that are in the top 10, but not summed. Slide does not have that large of a lead over the otehr applications out there. I commented on this on Alex's original post.

Comment from Bob Warfield on 2007-10-08

2007-10-09T03:24:24Z

No please, no filtering!!!!

Filtering is only going to make the long tail even less interesting. It will award all the bonus points to so few applications as to make building more applications pointless. It's already starting to look marginal.

Deepak is exactly right when he says your Friends are doing the filtering by deciding which apps to use.

If you can't trust your friends, who can you trust? More importantly, if you don't care about the choices your friends are making, why use a social network at all?

Cheers,

Comment from Phil on 2007-10-09

2007-10-09T07:43:00Z

What about the little guys though? This was a question to Dave Morin at Future of Web Apps, how do you stop big companies stealing ideas and ripping off smaller developers apps and using their size to push it out to many more people. Filtering or cutting down the number of apps will only serve to make it harder for the developers to do well or alternatively will destroy competition allowing for perhaps worse apps becoming more popular because they were the original?

I'd would, however, like to filter all references to vampires/zombies/pirates/werewolves/etc possible!