Blog Post

Twitter Needs to Become More Open or Die

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Twitter needs to become more decentralized and open, says the company’s former chief engineer, or it will eventually wither and die like other “walled garden” approaches to the web. In a blog post, Alex Payne says he quit the company at the height of its success earlier this year because he wanted the service to become an open, distributed communications platform, but the startup’s senior executives were more focused on building a business instead. It’s a critical question that many technology companies have faced: open or closed? Open can fuel more growth, but closed can generate more revenue. As Payne describes it:

Some time ago, I circulated a document internally with a straightforward thesis: Twitter needs to decentralize or it will die. Maybe not tomorrow, maybe not even in a decade, but it was (and, I think, remains) my belief that all communications media will inevitably be decentralized, and that all businesses who build walled gardens will eventually see them torn down.

The former Twitter engineer says he understands the company’s desire — and, in fact, its need — to develop a revenue-generating business rather than pursuing his vision of becoming a decentralized communications network, because “There are precious few case studies in the business textbooks of decentralization yielding substantial, predictable, sustainable profits for a commercial entity.” Although Payne doesn’t mention it, there is likely a considerable amount of pressure building on Twitter to generate revenues, if only because the company has raised over $100 million in financing, and its backers no doubt want to see some return on that investment.

However, the former head of Twitter’s platform team — who left to join a startup called BankSimple — says he’s come to agree with those who see Twitter as having a higher calling: namely, to become a decentralized communications network for the digital age, or what Om has called a messaging bus for the real-time web. We raised the issue of whether there needs to be more than one Twitter in a post earlier this year, in part as a result of the network’s repeated downtime issues, but also because the service has grown to become such a crucial news delivery system for many people. As Payne puts it:

The call for a decentralized Twitter speaks to deeper motives than profit: good engineering and social justice. Done right, a decentralized one-to-many communications mechanism could boast a resilience and efficiency that the current centralized Twitter does not. Decentralization isn’t just a better architecture, it’s an architecture that resists censorship and the corrupting influences of capital and marketing. At the very least, decentralization would make tweeting as fundamental and irrevocable a part of the Internet as email. Now that would be a triumph of humanity.

Among those who have called for a more open Twitter — or at least for the company to federate and inter-operate with other open services, such as the open-source network — is programming guru Dave Winer, the original developer of RSS and other web standards, who has written about the need for an open and distributed Twitter-style service and described how the company could integrate its service with others. Some developers of Twitter-based apps, including Jesse Stay of SocialToo, also appear to be thinking about their future, and whether open might be better.

In a way, this is a future that Twitter itself created by opening up its API in the first place, and creating a vibrant ecosystem of clients and services. Doing so undoubtedly accelerated the adoption of Twitter, but it also made the service seem less like a single company’s product and more like a distributed communications network like SMS or email. Now Twitter is trying to pull back some of that ecosystem under its own control, for its own business purposes, and in the process, it’s causing turmoil for developers (although Payne doesn’t show much sympathy for them in his post, saying they also need to decide whether they are building businesses or just fooling around).

The kind of open and decentralized network that Payne and others envision isn’t necessarily incompatible with generating revenue for Twitter the company. It would just have to compete by adding features or functionality to the core service, like everyone else — possibly by analyzing and filtering the data streaming through the network (although if it were truly open, others could do this as well). The core functionality of the network, the sending and receiving of messages, would effectively become a utility.

But the choice to embrace that future is ultimately Twitter’s to make, and for better or worse, it seems to have chosen the path of centralized control. Whether Alex Payne’s prediction about the company’s fate will come true remains to be seen. Twitter isn’t the only social network that is facing pressure to become more open: Facebook has also seen many of the same criticisms about being closed, and there’s been a lot of attention paid to potential open-source alternatives such as Diaspora, which recently released its source code. For more on the debate over “Open vs. Closed,” please see our earlier series of posts on the issue and some of the companies involved.

Related content from GigaOM Pro (sub req’d): Lessons From Twitter: How to Play Nice With Ecosystem Partners

Post and thumbnails courtesy of Flickr user macieklew

14 Responses to “Twitter Needs to Become More Open or Die”

  1. inboulder

    I think too many journalists forget, or don’t understand, that ‘twitter’ as a technology can be replicated in an afternoon, the ONLY thing that even remotely justifies the ludicrous 100mill investment is it’s network effect.


    Suggesting twitter take steps to decentralize is completely ridiculous.

  2. You completely misinterpreted Alex’s article. He didn’t leave because they wouldn’t listen to him. You just wrote this to get a bunch of hits. Such crappy, irresponsible work. As well, who knows. Maybe there will be a distributed network that emerges in China and the middle east as a way to promote free speech which then makes its way back here. Really disappointed in your article.

  3. Someone needs to start a little “Open or Die” website for this particular religious sect.

    Meanwhile, let’s see, AAPL’s walled garden is closing, today, around 276+. They surely can afford a helluva funeral.

    Followed by a boatload of very successful firms. Trolling down the River Styx.

  4. It is certainly a tough spot to be in, to have grown such a massive audience and opened up to so many developers, and now to be suddenly (probably forcefully) interested in generating revenue. Nevertheless, Twitter does need to partner or diversify, or become one of many clients who can accomplish simple social media objectives.

  5. “In a way, this is a future that Twitter itself created by opening up its API in the first place, and creating a vibrant ecosystem of clients and services.”

    Another “open” site, another API. How many API’s can web developers cope with? Maybe we should get some abstraction on top of that and create a “generic API” that will reduce the number of individual “concepts” developers must grasp.

    Model-driven development coupled with code generation can help in that. Domain-specific approaches are suitable (like ABSE, see to build that abstraction layer and make the translation betteen a concept and implementation code.

  6. Great post!

    Too bad the stake holders in all the big services physically cringed at the very word “decentralization” ( more so than even “open source” ) – they’ll regret it in the long run, I promise.

    If they could just stop with the ultra short sightedness of fiscal quarters and commit to long term profitability, they’d see there’s an order of a magnitude more money in decentralizing.

    Open beats closed, every time, without exception.

    Disclosure: I am a financial backer of Diaspora.

    • I disagree, Open does not always beats closed source. For the user it may be more versatile but for the company, it’s not a total win. Look at Linux, it has been open since it was formed. We have seen a lot of creativity and crapware I believe from it. In terms of business its not making a whole lot.

      On the other hand it is a different beast. It will be interesting with Diaspora a open competing platform against Facebook will be popular. If it becomes successful I believe an open Twitter would work. Otherwise it is a dream lost.

      • Sir, look back upon history.

        Time was, all the rail road companies had their own proprietary rail track gauge. One company’s train could not run on another’s. Some made a little profit, some didn’t.

        Then all the railroad’s decided to “decentralize” and use a standard rail gauge. Almost overnight commerce exponentially increased 1,000 fold. Good, services and passengers ( or “users”, if you will ) going all over the country. High volume sales, high volume profits.

        open > closed. *ALWAYS*


      • At the end of the day, it will be users that decide-not techno decisions about whether something is open or not.

        I bet that outside the open/closed debates that the IT world has, most users probably don’t care-they just want to communicate with their friends and share photos & videos and facebook makes that simple. They, the end user, doesn’t have to set anything up, the just log on and someone else worries about it.

  7. securityman

    It’s as if you didn’t even read the article.

    al3x writes:

    ” If you report anything in this post as a likely future direction for Twitter, you are a buffoon. Similarly, if you think this an attack on Twitter, you’re not paying attention. “