I like the part where he mentions:

“Having already built the back end, it’s a very small amount of effort to create any of these front end pages.”

It’s actually pretty easy to create different “views” on top of your data (see the way Rails or XSL can do this). But I think that the difficulty lies in testing and maintaining these different interfaces — especially if you are nit-picky about design and layout and how nice your brand looks everywhere.

Exactly.

Certainly the google solution is better than having to create a separate app for each of 11 different social networks. But still, what a nightmare..

It’ll be interesting to see if we start to see companies that start with a distribution point that isn’t the wild wild web and choose to expand into it later.

Amit

Not hell, really, but rather a market for startups. I’m actually comparing this to cellphones, where you have to adapt your game to fit in their tiny screens which vary from one brand to another.

It just a matter for days/weeks until somebody figures out the highest common denominator and just being a translator/adapter above all apis.

Gil

Gil, I think you’re right — and that’s just what Open Social seeks to do in a nutshell, provide that upper layer that translates down.

However, Naveen hits the nail on the head — there will always be gains to be had from customizing to a particular platform. And if one party starts doing it, everyone needs to in order to compete.

Look at the apps on Facebook — we all started by migrating existing web experiences onto the platform, but it very quickly went beyond that. Now every app is a soup of Facebook-specific code, and more thought goes into how to exploit the particular loopholes of the Facebook environment (vis-a-vis notifications, alerts, feeds, etc.) than goes into the behavior of the apps themselves!

Multiply that by 10 platforms and you have a big headache!