Opening up the Social Network Graph

LiveJournal founder Brad Fitzpatrick and Open Source Awards winner David Recordon just posted a manifesto titled “Thoughts on the Social Graph”. Brad and David presented their work at Foo Camp and have been sharing it with interested parties over the last couple of months.

Their project attempts to solve the problem of multiple overlapping social networks. This overlap makes it harder to establish new sites, as people tire of rebuilding networks on each social networking site. As a non-profit and opensource project, it aims to be vendor-neutral and usable by all vendors.

Brad sums it up:

Users and developers alike are going crazy. There’s too many social networks out there to keep track of. Developers want to make more, and users want to join more, but it’s all too much work to re-enter your friends and data. We need to lower the amount of pain for both users and developers and let a thousand new social applications bloom.

All I can say is: finally!

tags: , ,
  • http://twopieceset.blogspot.com Nick Gerner

    I’m not too familiar with the facebook developer platform, but would it be possible to build an fb app which grabs your social network and republishes it elsewhere somewhere open, standardized, machine readable?

    If so, this sounds like a fb app with lots of growth potential. Of course, fb might not like it so much…

  • http://tim.oreilly.com Tim O'Reilly

    When Brad talked about this work at the O’Reilly Radar Executive Briefing at Oscon, I was blown away. I’ve been talking for a while about the need for open source developers to focus on open data, but this is the first project to tackle a real open data problem, and in a unique, creative way.

    I think this project will go down in history as the first real open data project — and will be remembered in the same way as Stallman’s famous epiphany about the lack of source code to fix his printer.

  • Brad Fitzpatrick

    Nick: That’s been done already.

    And Facebook knows about it. They’ve heard my whole presentation, in person. While they might feel a bit weird about it, I think they know it’s inevitable, and it actually sound like they might want to participate in this project.

  • http://www.eflorenzano.com Eric Florenzano

    Tim: Have you ever run across MusicBrainz (http://musicbrainz.org/)? It is exactly what you’re talking about: open data. At its core, it’s a PostgreSQL database with an open license.

  • http://zooid.org/~vid davidm

    Sites like Facebook are just a reinvention of web “pages,” email, news. So this isn’t about integrating social networking sites, it’s really about integrating the internet. There are client (RSS readers, etc) and server (iGoogle) approaches to this.

  • http://therealmccrea.wordpress.com John McCrea

    This is a *very* important debate. No coporation, NGO, or benificent non-profit should own the social graph. Users should demand ownership, control, and portability of their data and content. My post on the topic is at: http://therealmccrea.wordpress.com/2007/08/18/the-great-debate-open-up-the-social-graph/

  • http://htp://freetube.110mb.com Live TV

    How about a social network that releases an API that lets you configure or modify ANYTHING within the network – complete transparency (except for user’s private data).

  • http://tim.oreilly.com Tim O'Reilly

    davidm –

    In some ways, facebook is “the web within the web,” but the friend network is an important new component. And when Brad got fed up with having to add his “friends” again and again, a frustration he’s probably shared with tens of millions of others, he actually decided to do something about it. And I think his approach will actually work.

    If sites adopt his APIs, the user will be in control of the social network. (Top tier sites may resist, but second tier sites will all be incented to join, so I think this is going to take over.) Even more importantly, the same tools will eventually tie into other, currently implicit, social network graphs. Who do I email with? Who do I IM with? Who do I call? I think Brad’s technology will ultimately solve the problem I wrote about in the post Social Network Fatigue and the Missing Web 2.0 Address Book.

  • http://groovie.org Ben Bangert

    OpenID 2.0’s Attribute Exchange protocol should help a lot not only with helping users join more networks, but avoiding duplicate entry of the same data.

    With MovableType 4.0 now having community style features and accepting OpenID logins (though I haven’t seen whether it supports additional OpenID 2.0 specs) the process should be a bit easier. Being able to manage your profile in one place and have the sites you joined be updated automatically is going to be a big boon for Attribute Exchange.

  • robotnik

    Will this allow me to use different social networks services for different groups of friends? eg, facebook for “personal friends”, linkedin for colleagues and business connections

  • http://www.craigsblog.com Craig Wood

    As a developer and a social network user I agree with all the statements above and am excited about a universal repository for a single repository of my social data. It is a chore to have every new community I join ask for my gmail address to find my friends.

    My concern is the ease for individuals who aren’t as tech savvy to leave personal information on one network and have it sucked down by a different less than above board network that is on OpenID Platform through Attribute Exchange.

    What is being done to make sure that data exchange is not only allowed, but that those who are receiving it have the best intentions? How will the community police itself? Is it completely transparent what data you allow shared through Attribute Exchange and is it simple enough for my Mother to understand?

  • http://www.snaptortoise.com George Mandis

    I think OpenID is the most straightforward answer to this problem at the moment. It may not solve all the problems but it’s a step in the right direction. If somebody like Google were to start participating in OpenID I could see it becoming much more widespread.

  • http://mediatrending.com Ryan Miller

    I’m all for openness, and love the idea of some sort of uber-control mechanism for all my social networking activities. I thought about it a lot when I was a Product Manager for Yahoo Address Book years back. At that time, email was the primary hub for social networking.

    Even as an avid user of Web 2.0 technology and social networks, I don’t use more than 2 networks with any frequency, and I don’t think the average person is trying to juggle and unite a lot of disparate social networks. Not only do I not mind keeping LinkedIn and Facebook separate (my two favorites), I *want* them separate. I’m in a completely different mindset when I visit LinkedIn (career networking) than when I visit Facebook (fun / interest-oriented). The UI, configurability, and privacy options of each are well suited for their purposes.

    Despite how enthusiastic the media/tech pundits are to talk about Facebook’s “platform”, I think the word “apps” is a misnomer. They’re apps from a developer standpoint because they are standalone creations created by discrete entities. But from the user’s standpoint, they are features of a web site. Facebook is a web site with a lot of features.

    I have about 250 or so people in my online social network. There is probably 15-20% overlap across my two favorites, LinkedIn and Facebook. The cost of having 2 social networks in my life is quite low, and there are advantages to walled gardens that supercede the cost of importing my Yahoo contacts two times vs. one time. I don’t have data to prove that I’m typical, but that’s my 2c.

  • http://wp.chickenandporn.com/category/tech/ Allan Clark

    Tim, Davidm, Brad-

    Open Data has existed for social networks, but the open network part is only a fraction of the “business case” or total offering to attract users and get the content populated to a marketable (users and adverts) resource. Many readers may focus on “that’s been done before” while ignoring that only 12 people used the iteration they’re thinking of.

    For example, FOAF is a format that describes us and our friends, and a bunch of OOB extras. FOAF files reachable via crawlers are an open social-network data-set, but FOAF is perhaps too-short: it provides a format, but no tools to use that. I can’t use my FOAF info to IM my friends, for example. A tool is required to spider the info and cache it, allowing us to use the data.

    Also, FOAF is too technical for most of my friends to even consider. “rdf? just… where do I click?” Being the friendliest guy on the block gets more Happy Hour invites.

  • http://groovie.org Ben Bangert

    Of course, its interesting that I hear some people yelling about how tired they are of re-creating their social network, while I see more and more social networks spring up with people joining in droves. Quite a few people happilly flock over to them should they have desirable features their existing network doesn’t have.

    Tony Stubblebine has made the rather convincing argument on his blog in the past that setting up a new account on many of these systems is not terribly difficult. As such, it is not a major impediment to gaining users, and if users are so tired of it then why do all these new social networks seem to get so many people using them? I have to agree with Tony, while geeks definitely are tired of making new profiles over and over, the average user doesn’t seem to care about it as much. Whether this is because they’re used to a certain amount of repetitive work, or maybe because not being a programmer, they don’t consider that there might be a programmatic way to deal with it.

    While OpenID definitely solves a real problem I hear many ‘average’ people talk about, none of the non-computer type people I know using MySpace, LinkedIn, and others are saying how horrible it is that they have to create a profile again when they join a new network.

  • http://flickr.com/photos/bootload Peter Renshaw

    “… while geeks definitely are tired of making new profiles over and over, the average user doesn’t seem to care about it as much …”

    @Ben Your average user might not care because your average user probably cannot visualise an alternative way of working. But it would be foolish to ignore a solution – a better way.

    A couple of years back I defined for myself how I want to work on the Web. It went something like this …

    “… I’m loyal to my data not applications,
    distributions or hardware. Chose your loyalties
    carefully. What is it that is important to you?
    Your data, applications that you use, or your
    image?

    … I have absolutely no loyalty to operating systems (OS). Nor do I care for applications. I don’t even have any particular preference to hardware. But data. That’s another whole other ball game. …”

    For me the only thing I really care about is my data, my ability to access and share it. In this it includes being free of silos – at the same time be able to take advantage of external Web-Apps, services.

    Tall order, sure. So any technical solution that helps me achieve this is worth further investigation.

  • http://www.yourfirstcreditcard.com Derrick

    Doesn’t this sound like the time when there were so many instant messaging networks until those unified IM programs came around and started allowing users to connect to multiple IM platforms?

    Well if this truly can be done, then I guess we all can save time registering and re-inviting freinds again.

  • http://www.stapleton-gray.com Ross Stapleton-Gray

    A bit tangential, but here’s an NSF award just made, re privacy in social networks:
    http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0713361

    “The goal of this research is to evaluate the privacy risks associated with information sharing in online social networks and to propose efficient mitigation strategies that may enhance privacy while preserving valuable online interaction. Online social networks such as Friendster, MySpace, or the Facebook have experienced exponential growth in membership in recent years. These networks are successful examples of computer-mediated social interaction. However, they also raise novel privacy concerns. This research will quantify the risks associated with information sharing in online social networks, increase awareness about those risks and ways to mitigate them, and therefore help increase the usability and value of those networks…”

  • http://danjgetting94@blogspot.com Jay

    I’m not convinced that the spread of social networks is as widespread in the “normal” population as it is amongst us “geeks”. Many people such as my wife are members of just one social network, and it works for her. What she wants is a simple user interface and to find that everyone else is on this network.

    BTW, anyone know how much FaceBook makes in advertising revenue per day?