The Twitter Tragedy
To become a successful company, it’ll have to kill the openness that makes it great.
Photograph by Kimihiro Hoshino/AFP/Getty Images.
It’s unusual for a corporate blog post filled with bureaucratic language to cause an Internet uproar, but that’s exactly what happened last week when Twitter published an item with this innocuous-sounding headline “Changes coming in Version 1.1 of the Twitter API.” None of the changes Twitter mentions, taken in isolation, seems like a big deal. But in the aggregate, they’re seen as an aggressive move against once-friendly collaborators, the third-party developers who use Twitter’s application programming interface to write apps like Tweetbot or Twitterific to access the service. Developers who write the apps and their friends in the community feel betrayed.
In its defense, Twitter’s only doing what it has to do to build a successful business. At the end of the day, that’s the real problem. Not every great idea is a great business, and the entire success of the Internet is built on the back of open technical standards that aren’t businesses at all—the protocols that make up email, the Web, and the rest—and to reach its full potential that’s what Twitter would have to be.
The Twitter fracas comes down to the fact that the company has a big problem. As a product, it’s a huge success with millions of users and tie-ins to everything under the sun, but as a business it has hardly any revenues. And yet keeping all those servers running is an expensive proposition. Venture capitalists have been willing to fund the operation since the basic service is so appealing and popular, but obviously at some point the goal is to earn a profit. And since Twitter has so far rejected the exit strategy of selling itself to a more established firm like Google, its investors will be looking for a big profit.
The most obvious approach would be more aggressive advertising, but here the third-party clients are a big problem. In Twitter’s earliest days, the service was primarily built around the idea of sending and receiving messages via SMS with a fairly crude Web interface as a fallback. But the service came with a wide open Application Programming Interface (API) that let people write apps for computers and smartphones that could access the Twitter stream. This capability is what made it possible for Twitter to scale up. Eventually the popular multi-platform client Tweetie was acquired by Twitter and became the “official” first-party client. The trouble here is that if third-party apps can access Twitter’s data and then display it whichever way they like, then stripping ads out seems like an obvious feature to offer. What’s more, unlike Twitter itself, the third-party developers actually do have a clear profit model—they sell software to users. But if Twitter wants to get into the paid app business, it stands to reason that they’d want to clear the field of competition.
So even though the screws are only tightening slightly, the tech savvy see it as a sign of things to come and are reacting with panic and outrage. Marco Arment says it’s “some serious bullshit on Twitter’s part,” while Dan Wineman says they’re “on the path of Prodigy and CompuServe.” The issue even expands beyond the narrow domain of third-party clients, with Tumblr cut off from access to friend-finding and speculation rampant that Flipboard will be barred soon.
The sense of betrayal is easy to understand. On the other hand, the basic reality of the status quo is that the more successful app vendors were making money and Twitter wasn’t.
And it’s not just app developers. Lots of people are deriving meaningful value from Twitter’s platform. Politicians use it to get their message out, big companies use Twitter as part of their branding strategy, Twitter’s been a key enabling technology of the food truck boom, and every Web publication on the planet uses Twitter to drive traffic. Everyone’s capturing value here except the people paying the bills to keep the service running. Twitter needs to do something to recapture that value.
One leading alternative is Dalton Caldwell’s new project App.net, which proposes to simply charge users $50 a year for a Twitter-esque service. As a business move, this—just like Twitter’s efforts to lock itself down—makes perfect sense. But from a broader perspective fracturing the audience into multiple rival platforms (Twitter, App.net, etc.) is destructive. Social media is valuable in part because of “network effects.” You want to use a service that everyone else is using. But diversity is also valuable, which is exactly what’s made the open API days of Twitter so utopian—it’s a single centralized service, with a potentially limitless set of entry points. It just doesn’t work financially.
On the other hand, you might ask yourself what’s email’s business model and how do the folks behind it ever expect to make money? The answer is that there isn’t one and they don’t. Email isn’t a business. It’s a protocol. Some businesses—Google, Yahoo, Microsoft—make money offering email accounts and then selling ads. Lots of other businesses offer emails accounts to their employees as a productivity tool. Telecommunications and Web hosting companies often bundle email with other services. But email as such isn’t anyone’s business—it’s a set of technical standards, part of the fundamental architecture of the computer landscape. If it were a business, it would be facing the precise Twitter problem—how do you make money while leaving the ecosystem open and flexible enough to meet everyone’s needs?
Unfortunately, the social networking space is essentially re-creating the short-lived era of walled-garden networks—CompuServe vs. Prodigy vs. AOL—that predated the mass popularity of the Web. But the Web, like email, is open standards rather than a business. Its main creator, Tim Berners-Lee, got a knighthood for his trouble and a tribute at the 2012 Olympics opening ceremony but didn’t become a billionaire or earn a return for his investors. It’s not particularly reasonable to expect Twitter or its investors to suddenly embrace utopian ideals, ideals that were surely easier to hold onto back in the day when nobody had realistic expectations of vast fortunes. But those ideals really are integral to the Internet we know today. The growing disappointment with Twitter doesn’t reflect a business error on its part, and it won’t be corrected by any competing new service with a different model. It’s an integral element to the idea of building basic utilities as businesses rather than open technical standards. And barring a vast cultural shift, it means we’re in for disappointment after disappointment as new technologies emerge.