RailsConf 2008: A Belated Look Back

Noel Rappin, Friday, June 13, 2008 @ 4:27 pm

This is about a week late, but I was out of the office, and blah, blah, blah...

This was my first RailsConf, and I had a great time along with a couple thousand of my fellow Rails buddies. The general feeling was positive, people seemed pleased to be there and excited about the future of the Rails platform.

Overall, the three themes of the conference were:

• Deployment Beyond Mongrel: Lots of discussion of new VMs, new server technologies and the like. Cloud technologies were very popular.
• Git, git, git: Not only was everybody curious about how the core Rails team felt about the move to GitHub, all the newest projects were on GitHub, and there were a couple of well-attended sessions on Git.
• $ $ $: There were several sessions about running a small Rails business or making money off your side project. One of these (Dan Benjamin's) was so well attended, that the convention center kicked half the group out because we were a fire hazard.

Some further thoughts:

• I started my RailsConf experience at a Birds of a Feather hosted by Pivotal Labs, where I got discussing view layers and my issues with Erector with the actual developers of Erector. That was a fun session.
• Joel Spolsky was the first Keynote, and he gave a very polished an funny talk that he's obviously given before. I was hoping he'd address his issues with Rails more directly, or at least give a talk with more Rails-specific content.
• DHH gave a very interesting talk. The first half of which was an exploration of why Rails maintains a productivity surplus. His reasons: Rails confessed commonality among applications, Rails ceded flexibility in favor of convention, Rails is built on great technology, Rails assumes programmers matter. Most of the rest of the talk was focused on suggesting things that developers can do to make sure they are using that productivity surplus to improve themselves so as to be best positioned to take advantage of the next great thing. He talked about 37 Signals 4-day work week and training programs.
• Jeremy Kemper gave a more technical talk about Rails 2.1. My favorite moment actually was before the talk, at breakfast. Somebody at my table was going on about how he wasn't sure anybody really needed to get up early to make the keynote or some such, unaware that the quiet and unassuming Jeremy Kemper was sitting next to him.
• Seeing Kent Beck was a personal highlight. His main talk -- just stories about XP, S and JUnit, and the like, was fun, but he was even better taking questions from the audience. I was most interested in his comments about the term "agile", which he clearly feels watered down XP significantly.

Other talks I really liked -- links are two slides where available.

• Stephen Midgley's talk on managing advanced search. Which was different from the way Ryan Bates suggested on the a recent Railscast -- both good methods.
• (Parenthetically, Ryan Bates was one of the leading rock stars of the conference -- every time I walked by him, somebody was coming up to him to tell him how great the Railscasts are. I didn't get my chance -- Ryan, they are really great...)
• Scott Chacon's session on Git apparently set some kind of record for having umpteen gazillion Keynote slides, in the service of animations showing off Git internals. When I walked in to the talk, I felt I was about 25% of the way to really getting Git. Now I think I'm about 55% there, not bad for an hour's work.
• Also really liked Josh Susser's talk comparing RSpec, Shoulda, and Test::Unit, and not just because his conclusions were similar to mine...
• The most fun talk was probably Obie Fernandez' talk showing the worst Rails code ever. Laughter and embarrassed recognition prevailed in the room.
• Chad Pytel gave a really nice talk with a step-by-step refactoring of ActiveRecord code.

Check out Gregg Pollack's video recap, too.

agile ajax

Comments

• Leave a comment
• Comment feed for this post
• TrackBack URL

None so far.

Leave a comment