Can #NewTwitter Swim Faster Than a Fail Whale?
The Coast Guard reported Thursday that a ship entering San Francisco Bay struck a 25-foot whale and dragged it all the way to Berth 57 at the Port of Oakland, where it was pronounced DOA. (I’m not joking.)
If you worked at San Francisco-based Twitter and you were looking for bad omens, this might be one. The company is famous for the “Fail Whale,” a cheerful little drawing of a whale being borne aloft by birds. The graphic appears on the Twitter website every time the messaging service is over capacity, which is distressingly often. So when real whales start impaling themselves on the cargo ships that steam through the bay a couple of miles away from Twitter’s Folsom Street headquarters, you have to wonder if it’s nature’s warped way of sending a protest.
On the whole, it’s been an exciting month for the venture-funded startup, which launched in 2006. The folks at Twitter, the company, have been working very hard lately to prove that they can do cool things with Twitter, the communications platform. On September 1, they introduced a slick new app for the Apple iPad and iPhone that’s truly among the best ways to access your Twitterstream on those devices. And now they’re in the process of introducing a new version of the Twitter website—naturally dubbed #NewTwitter by social media mavens—that mimics some of the nicest features of the Twitter iPad app. (It’s an interesting case of tablet design influencing Web design.)
But after spending a day playing with #NewTwitter, I still have my doubts about whether Twitter, the startup, can regain leadership when it comes to building innovative mobile and desktop applications that tap into Twitter, the microblogging service. The heart of the challenge for Twitter’s leaders, in my mind, is that the business and the platform are not the same thing. Thanks largely to decisions Twitter execs made early on about opening up the technology to outside developers, the service has taken on a life of its own, separate from any hopes the company may for commercializing it.
There are great applications like Flipboard and TweetDeck (I’ll say more about each of these in a moment) that let millions of people use Twitter, the service, every day without ever visiting Twitter.com or having a substantive interaction with Twitter, the company. There is a whole online marketplace for Twitter apps, oneforty.com, where you can browse more than 1,000 Twitter apps for the Web and hundreds for mobile and desktop platforms. But only a handful of these are made by Twitter itself.
What Twitter’s founders invented, in other words, wasn’t really a product. It was a new communications genre—one as useful in its own way as e-mail and text messaging, and just as important as a platform for experimentation. What’s bizarre, when you think about it, is the fact that all tweets still have to pass through Twitter’s servers. It’s as if AOL were handling every e-mail message in the world, or as if Verizon were handling every SMS message. It would be a completely absurd situation, if it weren’t for the fact that Twitter really did invent the idea of 140-character messages that get distributed to collections of followers, and thus has every right to profit from it.
The problem is that there’s something bigger at stake here than one company’s fortunes. In a previous column about Twitter, back in March 2009, I wrote that “If tweeting is truly fundamental—that is, if Internet users start to think of it as a basic feature of the Internet comparable to e-mail or instant messaging, as I believe many already do—then it may turn out to be too important to leave to Twitter.” I think that’s even more true now than it was last year, and it’s an issue that Twitter, its investors, and its users may have to resolve before either the company or the platform can grow to their full potential.
But enough pontificating for the moment. Let’s get down to brass tacks, and compare the Twitter apps being created by outside developers with those the company is building in-house. I think you have to conclude that the outside developers are still innovating faster—which eventually leads back to the question of whether Twitter should really be in the Twitter app business, or should simply be keeping the infrastructure running.
When you look at Twitter’s recent work, especially the iPad app and #NewTwitter, you have to admit that the startup is making important strides. In the past “the company didn’t quite know where it was going,” writes Alex Payne, who was an engineer at Twitter from 2007 to early 2010. Now “Twitter is getting focused,” he says. That’s clear from the iPad app, which has a beautiful scrolling stream of tweets down the middle and a nifty column that slides in from the right whenever a tweet includes a link to Web or multimedia content. It’s all a lot more…kinetic than anything we’ve seen before from Twitter.
Never mind that the app is a bit confusing to use, or that Twitter’s mobile developers seem to have made up most of their own navigation conventions and user-interface gestures. (The app is riddled with “non standard actions, panels, and interactions,” in the opinion of Brookline, MA-based mobile developer Greg Raiz.) At least Twitter saw the possibilities in the iPad, and decided to throw its own hat in the ring.
But for my money, there are more effective Twitter apps than Twitter for iPad. If you want to quickly scan incoming tweets from your friends, see who’s cited you in an @mention lately, or see who’s tweeted about specific search terms, there’s still no better tool than TweetDeck, which can show you up to three columns of tweets side-by-side. TweetDeck is slow and terribly crashy, but I like seeing all that information in one big array, especially the saved searches. (I’ve got a search set up just for tweets mentioning Xconomy, for example. You can view such saved searches in the Twitter for iPad app, but they’re buried two panels deep.)
Then there’s Flipboard, which turns your social-media feeds into a kind of interactive magazine. Flipboard can connect to your Facebook account and to a number of custom “sections” drawn from major publications. But at its heart this app is Twitter-centric, giving you a pleasant way to browse not just your friends’ tweets but the stuff they’re tweeting about. That’s a key point. If you follow me on Twitter, and I tweet about this column, then what will show up in the Twitter section of Flipboard is not the tweet but the column itself, or at least the first few paragraphs of it. (If you want to read the whole column, Flipboard will open it for you in a browser window, or send it to Instapaper.)
I absolutely love this function of Flipboard. The 367 people whom I currently follow on Twitter were carefully chosen because they tweet about content that’s useful to me. In this way, I vastly increase my information reach, learning about many things I would never have had time to discover on my own. True, I can find the same content through TweetDeck or even Twitter for iPad. But I have to dig harder for it: I have to read each tweet, decide from each 140-character clue whether the content it links to sounds enticing, click on the URL in the link, and wait for the linked content to come up in a browser panel. Flipboard does most of that work for me in advance.
But because Flipboard de-emphasizes actual tweets in favor of the content they point to, it’s an app that Twitter itself could never have built. The Twitter API—the wonderful open conduit that allows applications built by third-party developers to tap into Twitter users’ accounts as long as these apps are properly authorized—has given rise to a whole galaxy of apps that do cool and wholly unintended things like this. Most of these features are things that Twitter will never profit from.
There’s been a lot of attention paid in the blogosphere this week to a farewell blog post by Alex Payne, the aforementioned former Twitter engineer. For years, Payne ran the Twitter Platform, meaning he was the guy who decided how the API could be used. The API originally grew out of the Twitter website “as a means to enable outside developers to accomplish what the company, with its then-tiny and overburdened team, could not,” Payne writes. And a very good thing too—not only do we now have a marvelous array of tools that take advantage of Twitter, but the popularity of these tools kept Twitter itself alive and growing during that long, unfocused period when the company didn’t seem to be doing much innovating on its own.
Payne’s essay echoes my own feelings about the distinction between Twitter, the company, and Twitter, the service. In a nicely phrased sentence, he says “there’s an important difference between lowercase ‘t’ tweeting and uppercase ‘T’ Twitter, just as with democrat and Democrat.” Twitter may be a business, but tweeting is a medium—and over time, Payne says he argued to his colleagues inside Twitter, that medium should be decentralized, in the same way that long-distance telephony and instant messaging have been decentralized. Breaking up the Twitter infrastructure wouldn’t simply be the socially just thing to do, Payne reasons, but would make the system more reliable and more resistant to censorship and “the corrupting influences of capital and marketing.”
It’s probably not surprising that Payne lost this argument and left the company. Twitter needs to generate some kind of return on the $160 million it’s collected in venture financing. #NewTwitter and all of its features clearly represent a push to get Twitter users to spend more time at Twitter.com, which will mean more opportunities to show them ads and promotions. The Twitter site is growing into a “rich information discovery platform,” in Payne’s words, encouraging “the kind of deep exploration of the data within Twitter that has previously only been exposed in bits and pieces by third-party applications.”
But Twitter is late to its own party. Apps like Flipboard have exposed the data within Twitter, allowing exploration that’s even deeper (in Flipboard’s case) thanks to the fact that the app moves some of the stage mechanisms like tweets and URLs out of the way. And the third-party apps provide many important functions, such as URL shortening, that are still inexplicably missing from #NewTwitter. The rollout of #NewTwitter has brought Twitter some well-deserved buzz—but my bet is that as long as Twitter keeps its API open, professional third-party developers will create the greater volume and variety of Twitter apps, piloting the Twitter ship to new and exciting destinations. Perhaps Twitter should concentrate on keeping the engines running—and on shooing away the Fail Whale.
For a full list of my columns, check out the World Wide Wade Archive. You can also subscribe to the column via RSS or e-mail, and you can download Pixel Nation, an e-book version of the first 80 columns, as a free PDF file or a $4.99 Kindle edition.