Blog Post

We could build an open Twitter, but would anyone use it?

Stay on Top of Enterprise Technology Trends

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

Amid the recent brouhaha over Twitter’s future — which some say is aimed at restricting what developers can do with the real-time information network, in an attempt to monetize it more easily — a number of critics have proposed duplicating the network using open-source tools and principles. This idea, which has also been proposed in the past by blogging pioneer and programmer Dave Winer, seems to have a lot of merit: after all, if a short-messaging utility like Twitter is a useful service for society to have, then why not recreate it as an open-source project? The only problem is that others have tried to do exactly that, and have mostly failed to achieve any traction. For better or worse, we seem to be stuck with Twitter.

The latest kerfuffle started with a blog post from Twitter’s director of consumer product Michael Sippey, who said that the service plans to tighten the restrictions on use of its API by third-party developers — an announcement that came on the same day that Twitter shut down a partnership with LinkedIn that allowed users of that service to cross-post tweets to their LinkedIn feed. This led to a number of critical comments from outside developers about the company’s treatment of them, a relationship that has been somewhat strained in the past, as Twitter has tried to control more and more of its ecosystem.

Would an open Twitter be feasible?

Among those complaints was a proposal from developer Brent Simmons, the creator of a popular RSS news-reader called NetNewsWire and a co-founder of Sepia Labs, creator of an app called Glassboard. Although Simmons said he hasn’t been involved in developing a Twitter app, he said the increasing restrictions and tone that the company was taking would make him think twice about doing so — and if he did have one, he would try to get other Twitter app developers together to come up with a way of duplicating the company’s network so they could replace it with an open one:

I would get in touch with other client developers and start talking about a way to do what Twitter does but that doesn’t require Twitter itself (or any specific company or service). Once we came to a consensus, then we’d add support for whatever-it-is to our apps… And then we’d promote the new thing, encourage people to use it, help it grow. Then drop Twitter some day — or wait till Twitter cuts off our apps.

Simmons points out that the technical elements required for a short-messaging service like Twitter, in which users can “follow” each other to get updates pushed to them, aren’t all that complicated (although the company might argue that it’s a lot more complicated when you get to hundreds of millions of users and have to handle billions of simultaneous tweets every few days). A service that did this wouldn’t be all that different from the way that RSS operates as a news-distribution format, Simmons said, and a simple OPML file could be used to handle subscribing or unsubscribing from different people.

It’s no coincidence that Simmons mentions RSS and OPML as solutions to this problem: Dave Winer, who pioneered both technologies in the early days of the web, has been building a system that is based on those protocols for some time. Winer has written often about the need to reclaim the ability to publish short messages from Twitter’s corporate control — both because it would be better as an open service, and because it would be less likely to suffer from the kind of outages that took the network down in the early years of its life, when Winer proposed a kind of “emergency broadcast system.”

Twitter’s network effects are pretty powerful

But would an open Twitter have a hope of actually becoming an alternative to the real thing? Maybe two or three years ago something like that could have worked, but Twitter is now a massive network with over 100 million active users, and that’s a pretty powerful reason why people would tend to keep using the existing service. Not only that, but Twitter can and likely would do whatever it could to stop a competitor from emerging, just as it tried to stifle entrepreneur Bill Gross’s attempt to build a competing network through his company UberMedia.

In addition to Winer’s efforts, one company already tried to build an open-source version of Twitter: developed a client and service called, which was based on a model similar to that of the blogging platform WordPress (see disclosure below) — users could run the software on their own servers and connect to the network that way, or they could use a hosted version run by After a lack of uptake, apart from some die-hard programmers and the occasional celebrity, the company wound up pivoting to focus on a corporate information service similar to Yammer.

Diaspora, an open-source alternative to Facebook that was funded through a high-profile Kickstarter campaign in 2010, has suffered a somewhat similar fate: it has been criticized for not developing quickly enough, and seems to be used primarily by hobbyists, and others for whom the principle of an open network is more important than whether anyone else uses it or not. In the end, many users don’t really seem to care whether a system or network is open or not — or at least not enough of them to make a difference.

Disclosure: Automattic (maker of is backed by True Ventures, a venture capital firm that is an investor in the parent company of this blog, Giga Omni Media. Om Malik, founder of Giga Omni Media, is also a venture partner at True.

Post and thumbnail images courtesy of Flickr user Christian Scholz

17 Responses to “We could build an open Twitter, but would anyone use it?”

  1. joemaro

    For me StatusNet just works perfect and it does exactly what i want it to do. I am using the instance at
    Also Diaspora is working well and i’m hoping the planned update makes the federation a lot better…

  2. robertsteele

    OpenBTS (Range Networks, BurningMan cell service for 10,000) is out there, and if OpenBTS is possible, then Not Twitter is possible. Where we are all failing is in perpetuating the industrial era stove-pipe of isolated corrupt analytics. My new book, THE OPEN SOURCE EVERYTHING MANIFESTO: Transparency, Truth & Trust, addresses where we need to go now–right now–to break away. Microsoft, Oracle, Google, Facebook, Twitter, are all twisted manifestations of corrupt capitalism with rotten feedback loops (information pathologies. We need to cleanse the global information system, and in my view, OpenBTS and Open Non-Twitter is the way to start.

  3. Am equally as frustrated with the ongoing social networking wars, Twitter is just one of the culprits (blogged on this yesterday on allthingsanalytics dot com), however rather than seeing a single Twitter open source alternative, I would rather see many specialist networks pop-up, with aggregation clients on top that allow me to choose which networks I want to interact with – it might be pinterest for interiors, google+ for technology, twitter for entertainment, getsatisfaction for customer feedback, etc.

    Twitter is a noisy channel which tries to be everything to everyone, and that I expect will be its eventual downfall. But, if Twitter does start to go the way of HAL, then sign me up for an open source alternative ;-)

  4. Nick R Brown

    For 20 years users of the net have shown that they will happily jump off a bandwagon for a better bandwagon. Facebook and Twitter’s biggest enemy is buying into the notion that they are too big to fail. AOL, CompuServe, Myspace, etc, etc, etc thought that as well.

    • Anuj Agarwal

      Nick, I agree. 1% of twitter users tweet(create content), 10% curate and the rest 89% listen/consume tweets. Biggest problem for tweet consumers is the information overload. If a new product could solve this problem faced by 89% of twitter users(89% of 140million active users they claim = 124 million ) there is a chance they could win.

    • Wayne Spivak

      I agree Nick, and you can add to that list a plethora of technology companies from Novell to Kaypro, Digital Equipment, and Wang to Imation and thousands of software makers from Multi-mate to [fill in the blank].

      There is very little brand loyalty and that is what is driving the wheel. As you say, build a better mousetrap and they will go…

      Wayne Spivak
      SBA *

  5. Anuj Agarwal

    ‘Open Twitter’ (minus) ‘twitters Information overload problem’ should definitely work.

    Twitter is great for broadcasting but consuming tweets is a big big pain. Any alternative communication platform that could solve this problem should get mass adoption.

  6. Twitter clients provide a great backdoor to Twitter. The users of Twitter clients create a large amount of the traffic to Twitter and would be able to recreate a new ‘network’ themselves. Twitter knows this and has been buying companies that create clients, so they have the biggest part of these clients under their own control.

  7. I’m the cheerleader at buddycloud. The buddycloud team has been working to solve exactly the problems described in this article:
    1) federated by design – just like email, you choose your buddycloud provider and all servers find and talk to eachother.
    2) an open standard that anyone can adopt
    3) open source reference implmentation with a permissive licence (

    demo:[email protected] and[email protected]

    While Twitter suffers from the “marketers shouting at marketers” Tweet-and-forget, the buddycloud team has tried to foster a more conversational model akin to the threads one sees on G+.

    The “how will you reach critical mass” problem cited in the article is solved by growing out of topic channels – user follow topics and then connect with other users around each topic.

  8. Steffen Konrath

    I follow the idea of an open “Twitter” also for a while now. The service Twitter provides has much of a public good and should therefore also publically controlled/available rather than served by a company. The obvious problem is that we do not know how Twitter’s biz model will evolve over time. But I would go a different way as scribbled above.

    Steffen Konrath
    Future of Journalism/Media

  9. A Mitchell’s outcome doesn’t disprove the advantages of federated, open micro-blogging (OMB) platforms. Instead, it shows the risks of a developer-driven business model that under funds and under emphasizes the roles of marketing and keystone clients.

    Four aspects of made it difficult to use.

    First, failed to set out or enforce an anti-spam policy. Spam flooded the platform, driven in part by SEO advantages that (indexed by Google as a true blogging service) had over Twitter, which doesn’t provide much SEO juice. Power users could have helped self-police but were not provided opportunities for contributing.

    Second, readers receive all @ replies posted by everyone they follow, regardless of whether they follow the intended recipient. This is maddening and stupid.

    Third, with no built-in link shortener, is tedious and unpredictable as a Twitter client when URLs are involved.

    Last but not least, the people at never fully grasped the social aspects of microblogging. This made it difficult to find and connect with others and to use the OMB platform in an efficient manner.

    In the end, turned into a Twitter client for Iranians using it to avoid Tehran’s ban on direct access to Lacking localization and language-detection features,’s leaderboard is largely or entirely in Farsi. There’s no buzz or excitement for non-Farsi readers on any more. The brilliance and humor of the old community has disappeared.

  10. Dan Thornton

    The challenge isn\’t replicating the functionality of Twitter (or Facebook) in an open source environment – the challenge is really building adoption and critical mass, which is something few open or closed projects manage.

    I suspect and Diaspora both face the same problem, in that just being the open source alternative isn\’t enough to attract a mass of people, as the majority of non-developers don\’t really care about the code, or even privacy, compared to ease of use, and where the majority of their friends and family are.

    Having said that, if an open source project was combined with a decent USP for \’normal\’ users, there\’s no reason why it couldn\’t succeed – after all WordPress, Linux etc have all built decent or in the case of WordPress, amazing followings.

  11. I know this sounds terribly cheesy but I always assumed the mass public would insist Twitter become a public entity or something open would evolve into an open system. I guess it depends on how much the general public really wants and cares about a free and open flow of information using the technology that is available to us all.