You may remember the Facebook platform, launched in May 2007. “Right now, social networks are closed platforms, and today, we’re going to end that” proclaimed Mark Zuckerberg at the platform’s unveiling, which allowed external developers to build applications on top of the data Facebook had acquired abouts its users (Facebook calls this data “the social graph,” which has led many other startups to talk about how they owned “the XX graph” for their vertical.)
The euphoria was palpable as companies like Zynga and BuddyMedia leveraged the opportunity, acquiring millions of new users at relatively low costs. But then, Facebook started restricting what developers could do and slowly, the platform became less and less attractive to developers, with mobile replacing Facebook as the hot new thing.
According to figures from Social Bakers, about 33 per cent of countries on Facebook saw a decline in monthly active users over the last six months, compared to about 11 per cent over the last year. Global Web Index reported earlier this year that Twitter was now the fastests growing network, experiencing a 40% user growth rate in the last half of 2012 while Facebook was only experiencing a 35% one. And with many in the media talking about how the Facebook platform is in decline with developers (eg. Forbes, Wall Street Journal, AdAge, PandoDaily, etc…), a new opportunity has arisen and Twitter has quietly assembled all the pieces.
Quietly moving to a platform
First came the ability to embed tweets in other websites, an innocuous move that seemed to present no threat to a large platform like Facebook. Then came the follow button, which could be seen as the equivalent of Facebook’s Like button. This was followed by opening up the ability to leverage Twitter’s authentication framework, essentially providing login and user management services, to every developers who wanted to leverage it.
Of course, behind all this, Twitter has always been relatively generous in offering ways to get access to their data and augment it. Companies like DataSift have built strong businesses providing developers with access to all tweets coupled with extra social and sentiment data. And many startups were built and sold around the idea of providing independent software clients to manage how one managed his/her interaction with Twitter on mobile devices, connected TVs, and beyond. All of sudden, Twitter was everywhere.
On June 29, 2012, Twitter showed how it would build a better version of the Facebook platform. It didn’t come through a massive announcement on a stage in front of a lot of press but in a rather terse 439 words blog post on the company’s site by Michael Sippey, VP of product at the company. The post, entitled “Delivering a consistent Twitter experience,” highlighted Twitter’s ambitions in the application running space.
While much of the focus in the ensuing discussion about this post was relating to the fact that Twitter “developers should not build client apps that mimic or reproduce the mainstream Twitter consumer client experience,” a major portion of the post was dedicated to Twitter cards, with Sippey declaring “we want developers to be able to build applications that run within Tweets.”
Showing one’s cards
So what are Twitter cards?
At their core, Twitter cards are a way to display extra information that goes beyond the 140 characters self-imposed limit Twitter created. For example, here’s what a Twitter cards for last week’s entry looks like:
As you can see, in this case, a Twitter can include extra content including some images, information about where this was published, who wrote it and so on. Twitter offers different types of cards for apps, video/audio, photos, products, and more.
By allowing for extra information but not extra presentation to be handled, Twitter gets to control the user experience within the context of a card. This is important because it gives Twitter great amounts of control over how the content is being displayed but also allows them to present it in different ways on different platforms. So a developer can push out Twitter cards to Twitter and not worry about how they will be presented on the web, mobile, or TVs are Twitter will take care of that formatting.
And because cards allow for bite-sized content to be augmented, they fit neatly into the Twitter service, which has prided itself on artificial limits as to how much data it would carry. Interestingly enough, Twitter is not the only company pursuing cards as a new user interaction model.
Recently, Google unveiled Google Now. Google Now is a very pervasive system but what is interesting is that it presents its information as cards (even using the same nomenclature at Twitter) on mobile devices (see picture on left). For Google, the idea is that it can first leverage this information on Android devices, providing a clean and consistent way to deliver little bits of information. Expect cards to be coming soon to the search engine as it provides just the right amount of information around an individual content piece. As more and more digital device arise with smaller and smaller screens (eg. Google Glass, smartwatches), the ability to deliver small chunks of content becomes a substantial design issue and Twitter and Google are at the forefront of a major revolution in user interface design that will stay with us for a very long time.
Look, for example, at how Google is embedding cards into Google Glass, their wearable computing interface.
Just as the redesign of iOS7 has been on a flatter design template to focus on the content rather than on the interface, cards represent the logical next step, with most of the interface disappearing and the focus being put mostly on the content.
Twitter (and Google) are betting that this is the future of interface design, a future where minimalism wins over all and the context of the information drives the limited number of interface elements that are displayed. Because the interface disappears, it is the ideal model for any device, from feature phones or smartwatches presented with only a few lines of text to mobile devices, connected TVs or computers where you can use the enhanced data to augment the experience.
Bringing all the pieces together
Twitter has build up a large user population sharing different types of content; It’s also build a large following for its authentication service; It’s created ways for external parties to augment the data it has so as to match what Facebook does with its social graph; The ability to create micro-communities around a certain topic (using hashtags) has been embedded in the core messaging layer; it has location data for a large part of the tweets that are distributed on its network; and now it has a way to present all this on any device large or small.
If you start comparing the Facebook and Twitter platform, you end up with the following:
|Login and Registration||Yes||Yes|
|Portable UI||No||Yes (cards)|
|Share||Yes (Share)||Yes (Retweet)|
|Like||Yes (Like)||Yes (Favorite)|
|Social Graph Data|
|Apps||Yes (Ad API)||Yes (card)|
|Photos (& Galleries)||No||Yes (card)|
Looking at this chart, it’s pretty clear that Twitter is slowly filling the gap to matching all the capabilities of the Facebook platform. And don’t be surprised if they start unveiling products aimed at gather data around books and fitness or roll out a payment solution.
Where’s the backlash?
The Facebook platform was undone through a series of actions that led to substantial developer backlash. Because the company started from an “as open as we can possibly be” position and then proceeded to ratchet its openness as it figured out what business model it could operate on this platform, developers saw them as being overly greedy. By comparison, Twitter started with a relatively closed system and has progressively opened the door over time, sending out signals that is is getting more and more open over time. This is probably the main reason why we are not seeing the kind of negative press around the Twitter platform that we’ve seen about the Facebook one and it is a critical part of why the company may ultimately succeed where Facebook has apparently failed.