[imaginator]

Agree that it's shortsightedness.

But I think the problem goes much deeper: We keep building one-offs. Twitter - a one-off messaging service. Instagram - a one-off photo sharing app.

This is one of the reasons I started Buddycloud. We'd already built a nice location and social app but eschewed the VC cash to, dare I use the word, pivot, and build a different way of building apps.

Instead of building another one-off social-location-system like Foursquare, we decided it better to build a federated platform that others can then start building on. The federation and run-it-yourself mentality means that our users don't end up in the Twitpic scenario and that there are always other suppliers that will host your pictures in a compatible way. In a way that fosters competition between providers without needing to resort to switching friction cost to keep users.

In the Twitpic case, the buddycloud media server is designed to be a plug-in federated media hosting provider for each domain. Don't like how one provider is dealing with you data? Just switch.

It's not been easy to get this far, but we're starting to see traction from ex-app.net devs who are looking for something a bit more open that they can also run themselves (as a Docker container). And I still believe that the real solutions will be based on federated open systems that form a foundation so that developers can innovate further up the stack.