All of this is my opinion. I have been wrong before. Your mileage may vary.

A lot has been written in the last week about Twitter's API and developer concerns that it would soon be closed to apps that are platform-specific clients. The discussion was started by a post by a Twitter manager, Michael Sippey, who was also one of the earliest bloggers.

Then Brent Simmons, who I worked with for many years at UserLand, wrote a somewhat puzzling post (to me at least) saying that Twitter developers should start a migration to using RSS to connect their apps. That way, if Twitter outlawed clients, they'd have a means to connect their users.

That's the preamble. I have some comments of my own.

1. I thought Twitter had already said that developers should not do straight clients. Yes, this was unfair. Twitter had been telling developers, for years, that they should develop all kinds of clients. That was when Twitter was just a website and had no clients of its own. It bought a few of the developer products, and the advice changed. Really bad planning, and/or carelessness of the worst kind. I'm surprised there weren't any lawsuits. It was when they did this that I decided, for myself, that I would stop maintaining my Twitter apps, except the one I use to post to Twitter. I figured that functionality would be the last to be disconnected, if it ever was.

2. Smart developers will not just conclude that Twitter is unsafe to build on, but also any company that is operating in the Twitter model. If they are running a website, and trying to attract a lot of users, and are going in the direction of advertising, you'd be a fool to think they won't do the same as Twitter has. They just may not be as far along. I had an interesting conversation in Amsterdam last year with the founder of SoundCloud, who asked me to use his platform for podcasting. That's like asking a guy with a gambling addiction to put down a bet. I said no. At some point they will screw their users and developers as Twitter was already doing. I'll pay for my own hosting and use software I can run myself.

2a. Same advice for users re products like Feedburner. I think that was a bad bet and you'll eventually regret using it to serve your feeds. My own opinion.

2b. I have mixed feelings about companies like Dropbox. They charge money for their service, and that's a good sign. But they don't treat their customers like customers, enough to give comfort that they won't arbitrarily shut down developers.

3. What Brent is advising is exactly what I have been advising for a few years. Further, it's exactly what I've implemented in the combination of Radio2 and River2. I developed these products to play the role of a coral reef. I hope if anyone is getting started here, in bootstrapping an open alternative to Twitter, that they will be compatible with these two pieces of software. As Ben Franklin once said, "We must all hang together, or assuredly we will hang separately." This is a time to keep the variability to a minimum. However, I am not optimistic. Had Brent not published his missive, I probably wouldn't have said anything. I don't think Twitter developers create products to provide functionality to their users. I think they do it to provide an interface to Twitter. Perhaps a subtle distinction, but an important one. It's a lot like the days when I was an Apple developer. Everyone hoped they would be the one who Apple made a deal with. Few developers did what would have been smart for all of us, worked first with other developers to make each other strong. It's a variant of the Prisoner's Dilemma.

4. Speaking of Apple, why is it a bad idea to develop on the Twitter API, but a good idea to develop on Apple's? I don't get it. Apple is far bigger therefore far more dangerous than Twitter.

5. To news people -- pay attention. This is your future in the months and years to come. You may feel that Twitter is now a solid platform to build your business on. It is not.

Conclusion -- corporate APIs are good for the corporations that own them, and bad for everyone else. I would be reluctant to develop on any corporate API unless I was prepared to have my work completely deleted or obviated or usurped by the platform vendor. You really don't have any power. However it's impossible to avoid them. But try to. And don't be a crybaby when you get hurt.

Eventually the bubble will burst, and then we'll build on whatever open APIs we have, and the corporate APIs will explode along with the bubble. It has to happen, it always does. Until then, be smart.

 7/7/2012; 5:58:10 PM.