Monday, April 6, 2009

Why the arguments over FriendFeed beta's real time feed are ultimately meaningless

For those of my readers who haven't encountered the recent pleasantness/unpleasantness in the FriendFeed community, let me catch you up. Early Thursday evening, Louis Gray posted this tweet:

At FriendFeed HQ with @scobleizer @techcrunch @eldon @mattcutts @thomashawk @bhc3 @paultoo @btaylor and others, watching a demo.

Presumably it wasn't a demo of SimCity that lured this assortment of people to FriendFeed HQ on a Thursday evening. And after the demo, Robert Scoble tweeted:

Thank you friendfeed for verifying that I made the right choice. I am stunned and can't wait for Monday. Wow. http://ff.im/1TncR

By Monday morning, everyone was seeing what FriendFeed had cooked up, and the item that was getting the most attention was the default real-time feed in the beta version of FriendFeed. Some people loved it, while others literally became sick looking at it.

Now one way to evaluate the possible success of the new version of FriendFeed is to scan all of FriendFeed for mentions of the word beta (the beta FriendFeed lets you do that), take a look at the comments you see, and make a judgment from there.

With all due respect, that is the ABSOLUTE STUPIDEST thing that you can do. (And I mean that in a nice way.)

Think about it for a minute.

On Thursday night a group of experienced Web users, many of whom were very experienced in the use of FriendFeed, watched a demo.

On Monday morning the beta was open to all, but chances are that the comments you saw about the torrent of information in the real-time feed came from, again, experienced Web users, many of whom were very experienced in the use of FriendFeed.

So when you heard Robert Scoble share his experiences with the real-time feed, he was looking at a feed with over 14,000 subscriptions. Louis Gray was looking at over 8,000 subscriptions. Thomas Hawk, over 6,000. Michael Arrington, almost 1,400. Steven Hodson, who was teaching us all about Canadian pharmaceuticals, was looking at over 400 subscriptions. And I was looking at over 100 subscriptions. (Yes, I had over 1,000 subscriptions as ontarioemperor, but I'm building back up slowly.)

But allow me to let you in on a little secret. Existing FriendFeed users, you may want to sit down.

FriendFeed did not design the new interface for Robert Scoble, or Louis Gray, or Thomas Hawk, or Michael Arrington, or Steven Hodson, or me, or you.


FriendFeed designed the new interface for the millions upon millions of people who aren't using FriendFeed yet.

And guess what? New users don't subscribe to 100 people, or 1,000 people, or 10,000 people, or 100,000 people.

I'll give you an example. For all practical purposes, the empoprises FriendFeed account is a new account, so I like to monitor and see who subscribes to it. While many of the subscribers are these active FriendFeed users that knew me in the ontarioemperor days, occasionally I get a subscriber from outside of the Kool-Aid circle. In fact, one day I got a subscriber who only subscribes to one person - me. (The person is obviously deranged.) And there are probably a lot of FriendFeed users who only subscribe to a few people. (In fact, I bet that a significant number of FriendFeed users don't subscribe to anybody, strange as that may seem.)

Now I don't know the mean or median number of subscribers per FriendFeed user, but I bet that they are somewhat parallel to the mean or median number of people that a Twitter user follows. Now this February 2008 information is dated, but I bet it's still accurate.

[I]f you have about 10 followers and you're following about 10 people then you're Twittering away with a solid 50% of others like you using Twitter. If have more than 80 followers and you're following more than 70 people then you are in the Twitter minority—about 10%.

If these statistics also apply to FriendFeed, then all of the FriendFeed users that I mentioned above are VERY UNREPRESENTATIVE of the average FriendFeed experience.

If you're an existing FriendFeed user, take a moment and imagine what your real-time feed would look like if you subscribed to only 10 people. You know what? I bet it would look pretty cool. Rather than being overwhelming, it would give you the time and pace to read all of the information that was provided, and then every once in a while a new thing would magically pop up, just like that!

Now THAT'S the audience that the FriendFeed people want to target. And THAT'S why the real-time feed is the default in the beta. To have the real-time feed be turned OFF by default would be like...well, it would be like devoting all your energies to support users of Internet Explorer 6. (Heh.)
blog comments powered by Disqus