I spent most of last weekend hanging out at FOO Camp, the remarkable conference Tim O’Reilly hosts for 200+ geeks in the backyard of his corporate headquarters in Sebastapol, CA. Rather than try to transcribe the various sessions (which would likely require another 48 hours), I thought I’d offer the ideas that have stuck with me a few days later:
My new favorite toy, which I discovered at FOO, is del.icio.us, which appears to be the first online bookmark tool I’ll actually use. It’s convenient (runs as a pair of bookmarklets), flexible (able to put a page in an arbitrary number of folders) and socialble (you can see who else has bookmarked a page and then surf their bookmarks.) My bookmarks page is pretty pathetic at present, but I suspect it will become a form of linkblogging for me pretty soon.
Some of the best conversations of the conference concerned data visualization. Tim O’R and Roger Magoulas gave a talk about the data analysis they perform using Nielsen’s Book Scan and their internal sales data. I was less interested in their discoveries (dot.net rising, J2EE falling, the huge markets for books on consumer software, like PowerPoint) than I was in the tools they were using. Most of the research was presented in Treemap form, using a free Java library – I encountered Treemaps for the first time in Martin Wattenberg’s brilliant Map of the Market – I’m hoping to use the library in my GAP research soon. I also enjoyed some of the wacky ways folks are doing data visualization, bending old tools to new and nefarious purposes – is evidently playing with tools that look at data dependencies in software engineering to model the influence of money in US politics – very much looking forward to see what comes out of this. Also very happy to learn about gGobi, which looks like an amazingly powerful open source graphing and data discovery tool. I also had a great conversation with Scott Davis about mapping software, focusing on tools that can distort maps based on one or more factors – I just found a great example of one at the Electoral Vote Predictor earlier today.
My friend Quinn Norton gave a terrific talk on the economics of Kingdom of Loathing, an extremely goofy massively multiplayer online game that appears to break all the rules on how one should run an online game. I found Quinn’s observation that KoL is basically “a blog game” – the homegrown project of one very quirky guy with the input of lots of his users. Is this a new form of personal publishing for the extremely geeky? I suspect it might be – roughly 40% of all geeks I know have either tried to build a MMPOG, or have plans to build one. KoL is a great lesson that it’s not the graphics or the technology – it’s clever, geeky humor and a set of rules flexible enough to allow interesting behaviors to emerge.
On the subject of simple tools and complex behaviors – I spent a good chunk of my time at the conference listening to some of the discussions between the various blogging tool builders attending the conference. While one of my favorite blog tool developers wasn’t at the conference, many of the major developers were, and I had the sense more than once that, if someone wanted to make a huge change in how blogs worked, if one simply convinced everyone in the room, they’d be well on their way. A subtle, but very cool, possible from Kevin Marks of Technorati – we need a slightly more complex semantics of linking. When I add someone to my blogroll, it doesn’t mean that they said something interesting that day – it’s more a representation of our social relationship. Kevin suggests using a tagging standard called XFN, which distinguishes social links from content ones. He’s also looking for ways to add approval and disapproval to links: i.e., when I link to something now, I’m implicitly endorsing it – that’s problematic when you’re linking to something to critique it.
One of the best talks I attended was Brian McConnell’s on “Hacking Your Way Off the Grid”. Brian has spent the past few years tinkering with his San Francisco home, adding photovoltaic panels to cut electricity bills, solar water heat to his hot tub and solar air heat to certain rooms of the house. He’s tracked each step of the process and calculated how many years it will take to amortize each investment. My brain kept flying off in odd directions – could we use lightweight solar air heating units to warm gers in Central Asia? Can obsolete chip fab plants be repurposed to make PV cells? But Brian’s probably on the right track, thinking of simple changes that could make energy efficient technology cheaper and easier to install for US users, simplying the supply chain and applying lots of silicon valley ingenuity to making this technology accessible. His underlying point: what if US geeks spent as much time hacking and improving personal power systems as we do improving microcomputers? His Powerpoint is online here and very much worth reading.
Danny O’Brien’s talks are always worth listening to – I’m still improving my life by implementing suggestions from his “Tech Secrets of Overprolific Alpha Geeks” talk. This time, he focused on the future of fame, specifically the concept of microcelebrity. Danny’s a good example – if you hang around in certain (extremely geeky) scenes, he’s extremely well known. Outside of those circles, you probably haven’t a clue who he is. He suggests that’s the nature of this new medium. In an era of broadcast media, famous people were famous for everyone who watched TV – in an Internet era, it’s possible (and lots more likely) to be famous for a small group of people who care about what you do.
The most compelling idea he shared: We’re not able to scale our imaginations to accept the fact that there’s billions of people in the world, so we scale our world down to the people we know and care about. That’s why some of the best web links you will click on will open an entirely new world, big enough to have its own gossip columns, celebrities, society pages, etc. (For a good example of some of these links, read Biz Stone’s United Federation of Bloggers post, where he’s pointing to strange and wonderful blogs around the world.
But the most interesting idea at the conference was the conference itself. There’s less (apparent) structure for FOO than for any other conference I’ve ever attended. The OReillians designate certain areas of their office buildings as on or off limits, hire chefs to provide a light breakfast and two big meals a day, provide wifi and a wiki and a big board where people can propose sessions they’d like to lead. And that’s about it. And it works better than almost any other conference I’ve ever attended. I don’t know if the model scales, or if it only works with hand-picked attendees, or if there’s lots more structure behind the scenes that I didn’t see… but it’s remarkable and something that I hope others will learn from in putting together conferences, because it’s a great deal of fun. Many, many thanks to Tim O’Reilly and Sara Winge for making it happen.