platforms

I’ve been thinking a lot recently about platforms and services as platforms. The recent strategy trend in start up companies seems to be: become a platform.

That’s what Twitter is doing, that’s what Facebook and every other social network is doing. You build an infrastructure and an API and you invite people to play on your system.

But the one thing to remember is that the web is the ultimate platform.

I keep seeing the exodus of Facebook features, one by one, to the native web. Twitter is enhanced, open, archive-able “status updates.” FriendFeed is trying to co-opt the News Feed (what I give credit for popularizing the idea of the lifestream) and bring it out onto the open web. In many ways it is succeeding.

I think it’s inevitable that online community is going to end up as a disparate set of open services that work together instead of a closed system (Facebook) that offers all services.

The future of social networking is that everyone has:

• a blog (profile + notes)
• FriendFeed (news feed)
• Twitter (status)
• flickR (photos)
• del.icio.us or Google (shared items)
• etc., etc.

The smart way to go about “building a platform” is not to build something on top the web that traps users and developers, but to build something within the web so that it connects with everything that’s already available.

What’s the difference?

Facebook sits on top of the web, and it relies on its users and its developers to be content with only a base level of interaction with the greater web. When you build a Facebook application, you’re building a Facebook application, not a web application.

It’s the difference between wearing a Starship Troopers exoskeleton and working out. The exoskeleton is really cool looking, polished, and lets you plug in all kinds of gears and gizmos. But you’re not actually any stronger than you were. And your muscles aren’t really connected to it, even though it’s responding to your push.

Welcome to our community! If you like what you see, you may want to subscribe to our RSS feed!

Aspiring iPhone developers are getting rejection letters by the thousands. Meanwhile, developers who have made it into the beta program are reporting that the shallow integration leaves next to no opportunity to build anything meaningful. Update: Developer Craig Hockenberry has a very interesting perspective on one of the major issues at stake in the debate.

A few months ago at Community Next, Mitch Kapor was bemoaning a similar problem with regard to the Facebook platform. It appears that at first pass, most developers are going to be dissatisfied with the options available to them on any platform.

This is symptomatic of one of the fundamental tensions Kapor described in his talk: platforms want to open up slowly and test the waters. Developers want to do everything right away. In the end, smart people usually find a way to work around platform limitations and go on to build cool stuff.

That said, there is a right way and a wrong way to engage with developer communities, and it’s safe to say that Apple isn’t doing a great job. At the risk of driving everyone nuts with another Mark Zuckerberg / SXSW post, I’d like to hold the Facebook team up as an example of how to start doing things right.

After his much-lambasted keynote interview, Zuckerberg made the decision to appear at the Facebook developer garage the following afternoon. During the Q&A, I noticed that he was more at ease onstage than we’ve ever seen him. Perhaps it had something to do with the fact that he was in his element chatting informally with geeks.

Whatever the cause, he was positively ebullient — by Zuckerbergian standards, anyway. Afterwards, he made himself extremely accessible. And since then, he’s been giving out interviews left and right, which has earned him some — IMHO, undeserved — scorn from Valleywag.

If Facebook keeps this up, they’re going to become a poster child for how to interact with developers. Of course, I don’t expect Zuckerberg to pull a Steve Ballmer anytime soon. Running up and down a stage screaming “Developers! Developers! Developers! Developers!” doesn’t really strike me as his style.

But as we watched his eyes light up while answering our questions about the site he has built from the ground up, many of us in the room gained a deeper understanding of his passion for and commitment to the platform. As I told him afterwards, that was the moment when a many of us decided, “this Zuckerberg guy isn’t evil.”

At base, what this really means it that for platform owners to maximize developer engagement and productivity, they need to overcommunicate, demonstrate passion, and offer inspiration. That doesn’t mean that they need to tell developers what to do.

Zuckerberg — rightly, I think — demonstrated great reluctance to tell people what kinds of applications to build. But platform owners need to say, “Together, we are going to build something amazing! What’s more, it’s going to be fun! Developers of the world, to me! Now, here’s how this is going to work…”

Apple has every right to take baby steps with the nascent developer community forming around the iPhone. But in the process, they can make the road less bumpy for themselves by offering an unprecedented opening of the kimono.

I’m not holding my breath — it is Apple, after all. But even a statement as simple as, “we want to take it slowly because of x, y, and z,” would be sufficient.