It's all spinning wheels and self-doubt until the first pot of coffee.

LiveJournal, über social software

I've just noticed that LiveJournal added a School Directory, where you can search for schools you've attended and add them and the years you were there to your user info. Here's where I was between the years of 1989 and 1997:

Goodbye, Classmates.com? I've told people for years that every other social software app on the web is a reinvention of LiveJournal, done badly.

I'll never do it, but ever since I realized that LJ was Open Source, I've wanted to launch my own LJ-based social software site. The only difference would be that your User Info is what would appear as the main page under your personal URL, and the Journal would just be a secondary tab in the sidebar.

Because, after all, wouldn't that pretty much match 90-99% of the features offered by things like Friendster, Orkut, Tribe, and friends? (Oh, and if you use this idea, and make lots of money—hire me or something, will ya?)

Archived Comments

  • I tend to agree that LJ has some great software. Even behind the scenes, brad has some nice stuff going on.

    So, I have an LJ site going with just a couple of weblog/journals on it.

    I'm probably setting up another server for $WORK. God any more ideas you want to share?

  • Four or five years ago, I was trying to set up an LJ instance at work for general kibbitz and project knowledge capture. It caught on a bit, but I ended up getting a Movable Type install working with some hacks to integrate it with LDAP. For whatever reason, that seemed to catch on better. Either way, alongside a company wiki, it can be a cool resource if actually used

  • I'm not sure putting the User Info first, ahead of the journal, is going to work as well as LJ has.

    The reason LJ has worked so well, in my mind, is because the social networking stuff is there but is never the focus. It props up the other features of the site, instead being the end that the other features are merely the means to. Just as in life, the networking just happens as you go about the things you actually wanted to do.

    Interestingly, I've found that LJ is lousy place to find topical communities (unless you're into typical teenage popularity contests). It only works well to connect individuals, not collectives. It's very much like meatspace in these respects. (I'm not entirely sure where I'm going with this point, but it has some relevance to the other one that I can't quite grasp. Maybe you see something?)

  • "The only difference would be that your User Info is what would appear as the main page under your personal URL, and the Journal would just be a secondary tab in the sidebar."

    This is pretty much exactly what I'm doing with SporkID (work in progress). Except that I'm not trying to provide these social web services, rather, I'm trying to tie them all together and function as the glue between them.

    And while I quite agree that LJ blows the other blogging communities out of the water, SporkID's not prejudiced: it plays just as nicely with the other services and the independant blogs as it can. (Though obviously, it's limited by how open they decide to be.)

  • Aristotle: Oh, actually I'm being a bit snarky. I think most social sites are a disaster, and you're right—seems like the social stuff should be an enhancement to what you're really trying to do. Although, lately I've been liking the way 43{places,people,things} works.

    I'm surprised about the communities though... It's been awhile, but I had actually gotten some use out of them a few years ago on LJ. I figured that'd be one of the better features as time went on.

  • Bob, Does your SporkID have any similarities with OpenID? There's been some thinking about how to build some kind of inter-app integration atop OpenID, but no-one's really got beyond the chin-stroking stage.

  • I haven't looked at 43*---maybe I should.

    As far as the communities on LJ are concerned, they aren't entirely useless, but both SNR and activity levels are way low. Quite like the typical computer club / writer's workshop / keyboard school in meatspace, where the locality of people is a bigger constraint to the makeup of the membership than the intensity of their interest. On LJ, I think the key factor is that it's so easy to find dedicated topical online communities, which offer a far more satisfactory experience, that only those who cannot be bothered to look for them are compelled to look for corresponding communities on LJ.

    On a different note, another case of succesful social networking just occured to me---Flickr. There too the networking it is a very successful component, cherished by many, but again only a component that compounds the experience, not an end in itself. And arguably that's why it's so successful.

  • After some careful consideration, I decided to rename SporkID to something more generic sounding (Sporklet) and pull the identity stuff right out of the application.

    But to answer your question, absolutely, yes. I am definately planning on building on top of OpenID, and possibly Sxip (though not definately). I'm working on a modular reputation-based authorization system of identity, which I'm planning on setting up under a non-profit org in order to allay any fears people might have of, say, having my company get acquired by another larger company and then finding themselves having lost control of their identity.

    The general idea is, "Great, we've got a decent, easy to implement system of authenticating via OpenID, but the decentralized model doesn't work quite as well for authorization systems because systems can lie." So what I'm working on is a method of bringing different trust metrics under a single roof that's easy to access via a REST web services API. And then Sporklet (the stuff I described in my earlier comment) just makes use of the reputation/identity stuff in the background.