Dalton Caldwell has really stirred the pot about Twitter failing the Tim Berners-Lee vision of an open web by abandoning their platform strategy, failing to be data portable, and accepting ads as a business model. I wish him luck as he embarks on his journey with app.net to right those wrongs, but the truth is that Twitter had the right decision made for them – platforms are really hard.

I’m not saying that platforms are impossible. Can platforms be built? Yes. In fact Caldwell’s new project has a much better chance of succeeding than Twitter would have at the point at which they abandoned their strategy.

Why? Because a platform built on the foundations of a company already valued orders of magnitude more than the revenue it would generate as a platform would have doomed Twitter.

That is where Twitter had the decision made for them.

Why couldn’t Twitter have pulled it off? The life cycle of a platform is a unique one compared to other business types. The platform has to pre-date the applications who will build on top of it. It then has to maintain a leadership position in its space without being replaced by latecomers, copy cats, and vertical integrations. Finally, it has to be able to support a robust set of use cases once the market has matured without becoming irrelevant.

The first phase is incredibly difficult because few companies know that they should build a platform from the get go. By the time they’ve figured it out, the piece working against them, and I’m assuming that they are VC funded for this post, is equity economics. Why? The pivot most likely involves losing any revenue and firing most of the customers you’ve spent your previous equity signing up. Now you’re starting over, but you’ve gotta convince folks to take C-round equity for an A-round company.

Luckily for app.net, Twitter has already done that work for them. They’ve demonstrated a market exists, and created the demand for an open solution. At the time Dick Costello took over at Twitter, they had taken hundreds of millions in funding, and there was no way committing to the no-revenue (or we’ll figure it out) business model was going to be a viable option.

The second phase is a tough one. Product market fit as a general solution is difficult. Folks showing up solving a vertical within the space are going to solve it better than you within the vertical. Copy-cats will show up trying to solve the general solution better given what they’ve seen you do wrong. Google is a great example. They compete simultaneously with Kayak and DuckDuckGo. Without some defensible asset, a platform will be run over by the gold rush to take over the new white space created.

I think app.net is starting in this phase, and needs to prove that there is enough white space in the form of customer demand. I know that the tech ecosystem is excited by data portability, partner incentives, and the like, but does Oprah care enough to get 25 million people to sign up for it?

I’ve never lived the third phase, but I imagine the folks at Adobe are trying to figure out what they did wrong in this regard. With YouTube, they were on top of the world. The Internet needed Flash, and Adobe was going to be the king of the roost. Then, all of a sudden, with the iPhone, Flash became the ugly step-child of the web. And just like that you’re irrelevant.

The app.net stack is a long ways off from that, and I hope they get there, but it will be interesting to see what happens in the meantime.

Good luck Dalton Caldwell, and the app.net team, but at this point in the game your best bet is to thank Twitter for paving the way, and focus on taking over the white space they carved out. Oh, and Twitter definitely made the right choice for themselves.