Right now, the big word on the internet is an API. People think, if we have an API, then we will have a platform. Once people start using our API, then they will be sticky onto our system. This is all true and is why API’s are great.

But how can you make your system truly sticky?

Instead of releasing just an API, you should make your system extensible. What do I mean by this? Look at Wordpress as a good example. Wordpress has already created the foundation and anything you would like to add to it, you can by only changing the pieces, instead of the whole.

In many cases, even in B2B cases, someone wants to use the API not because the existing one is bad, but because there is some functionality that works well for them which may be too niche for anyone else. With an API, the user would be expected to build their whole system from scratch. With extensibility, the user only has to change the pieces they do not like. If I wanted to create a pop-up ad appear everytime I load my main page, its an easy add to the code. If I wanted a random audio file to play with each post, again, its a simple changing of that page.

When you are building a platform, this is doubly important. Every company runs differently, they have different cultures, different workflows, when they decide to use your platform, they want to keep all of those things.

The big news today was Bebo creating their own language which mimics the facebook application FBML. The idea was essentially to allow developers to easily port from a Facebook app to a Bebo app. Everyone is talking about it in terms of the Bebo side, but I wonder whether this could be the first tentative step for Facebook to solve the Death problem.

Internet history (all 10 years of it) has shown time and time again that social networking sites do not survive very well. We have seen the coming and going of Friendster and MySpace. There was also pseudo social sites like Xanga and AsianAvenue. Users have shown time and time again that they will happily jump ship and in relatively short amounts of time. So, what is the solution to live?

The answer is, become a platform. I don’t just mean facebook apps, I mean, become the base platform in which everyone else runs. I mean a company like Ning. By allowing others to build the social network, everyone wins. As different networks become popular, Ning will be around as the backend, for users, it means no need to recreate profiles, even for individual networks, it means they have ready access to a large pool of possible users. It streamlines the process of life and death of networks. The main problem that Ning runs into right now, which is where Facebook currently has the great advantage is in sheer size. Facebook already has a large user base, one which social networking companies would give money hand over fist to build off of. If Facebooks competitors all were building their platform off of Facebook’s data, they will always have the advantage. This is in fact what Ning is hoping to build off of.

Still haven’t completely composed all my thoughts on this yet, maybe more to come.

Here is my guess though, if Facebook does not open up more, a lot more than just helping to build a mimic to FBML, you will see Ning (or competitor) begin to win in the coming years.

UPDATE: HERE is someone explaining what I’m thinking in much better words than I can.  Though from the Bebo point of view as opposed to Facebooks.