There’s been a lot of buzz in the Twittersphere this morning – well, when Twitter behaves, that is – about a new Twitter-like service, Identi.ca. Launched yesterday, Identi.ca takes the Twitter idea and open sources it, and that alone makes it worth keeping an eye on not just because we can soon expect a world full of Identi.ca installations, but also because it means that business will be able to take the code and run it behind the firewall, finally bringing Twitter-like ambient intimacy to enterprise. (If any businesses are brave enough to experiment, that is!)
In my view, though, being open source isn’t going to be enough of a draw for most people. Even if you assume that the service will turn out to be stable, reliable, richly featured, able to easily import contact lists, and attracts the interest of third party clients like Twitterific and Twhirl, that still won’t be enough to draw people away from Twitter, unless Twitter catastrophically fails. Yes, Twitter’s having significant and annoying problems, but it’s important not to underestimate just how apathetic users can be when it comes to migrating from one social system to another.
What Identi.ca needs to do is to become a cross between Ping.fm, which allows you to posts to multiple social networks, and FriendFeed, which aggregates your output from a variety of tools such as Twitter, Flickr, and Del.icio.us… but with bells on. We need a ‘write once, post anywhere’ system, combined with an ‘aggregate and de-dupe’ system, so that we can all become tool agnostic. Such a system wouldn’t care where you wrote your update, it would distribute it to all the tools you use, and it would aggregate back responses from all your friends, regardless of which system they used at the time.
I think there are two key parts to such a service: De-duping will be essential if such a system is going to be at all usable. If you post the same message to Twitter, FriendFeed, Plurk and Jaiku, then I don’t want to see it showing up four times in my aggregated feed. Friend list management and grouping is going to be the other key issues. The tedious thing about Identi.ca – or any other such service – is recreating my Twitter friend list, or at least some part of the Twitter/Identi.ca friend lists Venn diagram. This is possibly where something like OpenSocial might come in very handy.
I doubt that such a tool would be simple, and relying on other people’s APIs creates multiple points of failure, but the nice thing would be that if I am posting to all ‘microblogging’ platforms and aggregating them all back again, it won’t matter if one tool goes down for a bit. If Twitter dies, but my update has gone to FriendFeed too, and then pushed back out to my friend’s account that they happen to access via Jaiku, who cares that one route in that network was out of action for a bit? On the other hand, if Identi.ca were to becomes that WOPA-AADD system, then you are rather creating a single point of failure… unless, of course, people were to run multiple installations as nodes in a Skype-like network, which would be possible with open source code. Just a thought.
Whilst I doubt that I’ll be deserting Twitter any time soon, if Identi.ca moves in the right direction it could really make a big difference to how we maintain our online presence.