onfocus

Just for some OSCON closure...I enjoyed the rest of the day. Michael's talk convinced me that I should be looking into this whole "Ruby on Rails" thing, it's impressive what he's put together (llor.nu, unroll backwards) with Rails and Scriptaculous. He mentioned that he barely knew JavaScript, but was able to put together the highly interactive interface for the game thanks to existing libraries.

I saw a demo of the Google Ajax Search API, and got a better feel for that. It's definitely buzzword compliant, but I'm not convinced it improves over the standard Google API beyond the perceived response time. I personally wish Google would put as much effort into expanding their existing Search APIs, rather than delivering the same data in a new, souped-up way.

The rest of the day I was on the "Six Apart" track. Artur Bergman lead a session about 6A's infrastructure. Lots of talk about load-balancing, caching, and managing servers ensued. Then 6A-er Tatsuhiko Miyagawa demoed his project Plagger—an RSS/Atom slicer and dicer. It does a ton of cool stuff, but it looks hard to install. I need to play around with it.

I didn't make it to the community sessions I was talking about, but I had fun peering into windows on tech worlds outside of my normal experience.

I'm blogging live from the Portland Convention Center where OSCON 2006 is well underway. After attending Webvisions last week, it's amazing to see a conference that is several orders of magnitude bigger. Instead of a few hundred people spanning a handful of conference rooms, OSCON feels like thousands of people spanning dozens of rooms. (Don't quote me on the numbers.)

It's always great to hear Tim O'Reilly talk about what he's thinking about. He listed five ideas related to open source that are designed to provoke the audience:

1. Architecture of Participation
2. Open Source Licenses Are Obsolete
3. Asymmetric Competition
4. Operations as Advantage
5. End of Open Data

Out of context, these probably don't mean much. But I've added them here for my own benefit. And he included a couple of quotes that I'd like to remember: "When the best leader leads, the people say, 'We did it ourselves.'" -Lao Tzu, and "In the future, 'being on someone's platform' will mean being hosted on their infrastructure." - Debra Chrapaty, VP Windows Live.

Anil had a keynote this morning as well, and he talked about making meaningful applications to help people connect. He also talked about the open source Six Apart developer tools for building scalable web apps. He mentioned that most of the applications we think about when we think Web 2.0 are using at least some of the tools. I didn't realize how widely they were used.

Conference organizer Nat talked about a new focus on talks about community at OSCON—one of my primary areas of interest. And there are a bunch of sessions today that aren't specifically about coding that I'd like to see.

I'm in Michael Buffington's talk about games with Rails now...

I spent all day yesterday at OSCON in Portland. The day started with a keynote by Freeman Dyson and his son George Dyson (moderated by Tim O'Reilly).



One of the things that struck me from the conversation was the idea of the "domestication" of technology. Freeman felt that the failure of nuclear technology was in part due to the fact that you can't have a small nuclear project. In other words, you can't run down to Radio Shack and pick up a fission kit and power your home projects with nuclear energy—it's solely the domain of large projects. By contrast, biotechnology is becoming domesticated. Freeman mentioned plant and animal gene-splicing kits for backyard breeders that are only a few years away. He mentioned a future children's game where kids compete to see who can grow the prickliest cactus, and the fact that DNA synthesizers—while currently outrageously expensive—are coming down in price. What I took away from this is that decentralization and adoption by a wide number of people is a key attribute if a technology is ultimately going to be successful.

There were a lot of other good elements in the talk, and they did discuss the dangers and unintended consequences related to new technology. Though Freeman said luck would prevent catastrophe, as it always does. (hmm.)



For the rest of the day I was on the O'Reilly-track. I saw a preview of their new magazine, Make, which looks fantastic. It's produced by Dale Dougherty and edited by boingboing pioneer Mark Frauenfelder who were both on hand to describe it. One of the first feature articles is how to build your own kite photography rig. Dale mentioned the inspiration for the magazine came when he realized there was no Martha Stewart for tech geeks.

I also saw a demo of SafariU, a tool that lets teachers assemble custom books from various sources. This is definitely a disruptive technology for the college book market, but should be very appealing to professors who want more control over course material. And I gave a demo of the newly launched Safari Affiliate Program, and their related Web Services—a project I've been involved with for a while. Like I mentioned yesterday, Safari is doing the important work of making books available as bits in addition to atoms—something I first read about in Being Digital almost ten years ago. It seemed like a far-off future at the time, but here we are.