On fully distributing the social network

by

Justin and I have been talking a lot lately about what’s wrong with social networking. Much has been written about social network fatigue and about the lack of data portability provided by many of the major social networks. For a variety of reasons, the portability of my identity and the graph of who my friends are and my relationships to them – in other words, me and my social network – is an extremely interesting problem to me. (And as Eric says, every good piece of software starts with a developer scratching his own itch.) Perhaps I’m not so interested in data portability aspects of getting my photos out of Flickr or my bookmarks out of Delicious because it’s already so easy to do. Getting my information about myself and my social network out of Facebook isn’t easy to do…

Facebook’s approach is a classic old-fashioned business model propped up by creating artificial scarcity where none actually exists. It’s much like the problems with the academic publishing businesses right now. The journal publishers want me to come up with a great research idea, go find funding for the work, do the work, write up the work, and then completely sign over all the rights to my work to them – so that I have to pay a license fee to use my own writing with my own students in my own classroom. Facebook wants me to have meet lots of people, make friends with many of them, spend my time connecting the dots between myself and my friends online, and label our relationships, so that Facebook can tell me I don’t have permission to use my own work. Springer, Elsevier, and Facebook… just another couple of data silos.

So Justin and I asked each other, instead of making it easier to get our data out of those silos, why trap our data in those silos in the first place? Let’s just bypass the whole problem. In the same way that we publish our research results on our own blogs instead of having the results hijacked by (published in) peer-reviewed journals, let’s take the same approach with our identities and social networks.

This would have to get implemented somehow, and since Justin and I are both WordPress users here’s what we’re thinking:

  • There doesn’t need to be a social network to join – you don’t need to pour your data into yet another silo.
  • Your personal blog is the perfect place for all your profile information (a la Facebook) and your identity information (including the kinds of stuff you would aggregate and share via FriendFeed) to be aggregated and displayed. It’s your site; why not keep all your information about yourself and your friends there?
  • Existing blog mechanisms like trackback and trackback moderation already show us how we can set up an “Add a Friend” feature in which we can build up a list of friends annotated with relationships (aka a social network) and then expose this list as XFN, FOAF, and whatever else you like (seems like RDF would be a natural choice).
  • Existing plugins let us replicate most of Facebook and similar sites’ functionalities, including extended profiles, a “mini-feed” of what your friends are doing, the ability to “poke” friends, and of course the plugin architecture gives you an open platform for extending core functionality (replicating Facebook isn’t the end goal, but if we can’t “at least” do that it will be hard to get traction)
  • Additional plugins could take us quickly beyond what Facebook and other sites have to offer
  • Once your identity becomes completely intertwingled with your blog, things like OpenID start to make much more sense – your username is now also the URL to your identity
  • &c.

This no-silos / everyone owns their own data approach gives you a fully distributed social network. WordPress-Multiuser and existing plugins give you 70% of the fully distributed network. Let’s do the last 30%! Let’s not just “open” social networks, let’s fully distribute them and take back control. Let’s not just demand permissions to use our own data, let’s just own our own data so that we don’t need anyone’s permission.

Justin is blogging some of the other people thinking this way. I know Brian and Jim and Darcy are keen on extending WP-MU to support more personal, intimate learning experiences. Who else out there is interested?

6 thoughts on “On fully distributing the social network”

  1. Yeah, the home page as home page. Get the content from _me_ don’t make me dish it out into your silos.

    You’re spot on in wanting more and more WPmu/WP widgets and plug-ins to make the site more inclusive. Vox.com does a lot of this pretty well, but then ends up failing the user on a number of the more basic functionalities and features.

    But re. OpenID does this mean that self-hosted blogs must be OpenID providers?

  2. I have been doing quite a bit of thinking on this very issue for quite some time now. I have been wondering why it is that we decentralize our identities (not online or offline, just our identities) the way we do. I post pictures to flickr, status updates to twitter and FB, keep a running reading list at del.icio.us, but I haven’t been able to take the next step and bring it all together around me. What I mean is that the only space that I post stuff to is my domain. It is truly the only place that I have the power to maintain long term.

    I think the idea of maintaing your identity in a central location is the long-term goal. I would prefer to use my identity to assert who I am against services instead of recreating identity at each. One of the things that bothers me about the current model is the need to recreate my social network each time I explore a new environment … as an example, just the other day I wanted to try out Pownce and to make it meaningful, I had to beg a handful of friends to join and add me. If the model were different, my identity would be asserted with Pownce, my network would come with me, and the only new thing is that my identity would now contain an attribute that let me into Pownce. Does that make any sense?

    At any rate, a few weeks ago on the ETS Talk podcast we discussed this topic and tossed our the idea of trying a 30 test — to se if we could stop with all the other social networks and use only our blogs as the center piece to our online identities … I haven’t gone down the path yet, but I reading your last couple of posts has given me a nudge in that direction once again. At any rate, I am watching what you are doing closely … I think you are getting close to the right approach.

  3. I just ran into Plaxo Pulse. Yes, it’s a social networking site, so it’s not quite what you’re describing, but it’s a lot less silo-like than Facebook – basically you can just add feeds and info from all the sites you do use (your blog, twitter, jaiku, flickr, etc.) and the feeds get collected and become your life stream on pulse. Also, you can get an rss feed from your network’s feed, so you can follow them with whatever reader you like, instead of looking at ads on Facebook. And you can remove your info as well, if you want to.

    The similarity to what you’re describing is that there’s practically nothing stored in Plaxo, except maybe the friend network – all content and submissions come from other sites that people use. Time will tell what happens when OpenSocial widgets start appearing in Plaxo…

  5. This is an excellent idea, and one of the main thrusts behind the new BuddyPress project – a social network built on WordPress MU. However rather than release BuddyPress so users can run wild creating their own mini-silos, we want to free the data. RSS, RDF, XFN, FOAF, OpenID, OAuth and more are the cogs – the machine we’ll build to make them work together is going to look pretty amazing.

    One thing I’m particularly interested in is synchronisation of profiles across multiple blogs and social networks. The fact is people have lots of different interests, and while they *may* have a central place to manage their data from, it’s also likely they see several of their online “homes” as important as each other. A decentralised list of ‘rel=”me”‘ places could all talk to each other, so if you update your “About me” text on one site you could choose to sync it to all your other profiles. Same with user statuses, if you want to say you’ve just discovered the perfect sandwich you don’t want to have to type that several times in several places.

    A lot more thought needs putting into that, but the potential for this whole area is very exciting.

  6. This was written 5 years ago. Has anything come of it?

    I want to be able to do the social network thing without having the social network involved.

Comments are closed.