Messaging. Federated? Silo? Does it Matter with an API and Bots?

January 19, 2016

Nobody cares anymore.

Who Cares
Nobody cares if you are a silo or federated as long as you’ve got an API

It used to be important. Interoperability. Federation. Communication across networks, devices and vendors. All useless now.

We’ve got our lowest common denominator: IP, HTTP – the web. We have our point of federation/aggregation – they now call it the home screen of a smartphone.

People got all riled up on my blog last week something about Wire needing to federate – check the comments. My view? Federating wouldn’t move an inch in their user’s base needle.

Today’s openness and messaging is all about being the platform and enabling others to connect to you. How is that achieved? By way of APIs. And by this new stupid word – “Bots” – which most probably stands for automation.

Why is Bots stupid? Because it just means using the API in a certain fashion.

Back to Messaging.


If you have a service. What happens if it is a silo?

You gain users to it. Slowly or faster. Doesn’t matter that much (though it probably does to you).

One day you want to add more utility to the service – some stickiness – making sure people don’t leave. So you add features. But you understand at some point that going it alone won’t move you fast enough, so you open it up to external developers and services.

You publish an open API.


Federation you say? You make your service accessible to other networks by interacting with them using the same protocol?


But what does that give you exactly? Same set of features you have, give or take a feature or two. Same utility. No stickiness. No differentiation. Not enough.

So you again wish to add features, but getting out of the core you’ve federated just places you in the position of proprietary features. And again you’ll one day understand it isn’t enough. Faster “innovation” and growth are required on that front – you can’t cater all of your customers’ needs. So you… open up! Slap an API on top of your service.

No one cares

Two alternatives. Same end result.

Time to move on. Nothing here to see.

Just make sure you have a solid infrastructure – and an API on top of it for others to integrate with.


You may also like

Two years of WebRTC Insights

Two years of WebRTC Insights

Your email address will not be published. Required fields are marked

  1. closed ecosystems are great for expanding to a proprietary user base, would question that if this is really sustainable. what’s becoming apparent is that consumers are wanting to free themselves of mundane activities, bots (automation) will require a great amount of third party development to solve the immense hurdle out there, particularly as those mundane tasks become more than 2-3 steps. This then means will any single one ecosystem really be able to solve for that or will it be a multitude of solutions, looking at apps they have evolved to single purpose solutions in large part because apps do 1-to-few things really well. Bots/automation solutions coming out would be wise to try to solve a collection of activities but operate under a framework that is interoperable so that the broader ecosystem thrives as consumers see the overall hive of automation come to reality.

    1. Pratap,

      I would have agreed with this statement if reality wasn’t so much stacked against it. A few closed ecosystems of the top of my head: Apple App Store, WhatsApp, Facebook, WeChat, Snapchat. All huge, sustainable and profitable.

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}