RailsConf 2007 is almost here and I haven’t had a chance to write anything about it yet, which is bad, since I’m giving two presentations there this year. It will be interesting to see how things have grown since the small conference in Chicago last year. The speaker list for Portland is impressive and I’m looking forward to quite a few of the sessions and keynotes, Avi Bryant’s keynote. I’ve been following Seaside and DabbleDB a little bit and look forward to hearing more about his continuation based framework.

My two sessions are the afternoon of May 18:

Rails in Higher Education

1:50—2:40pm, Oregon Ballroom 201

The University of Nebraska, Lincoln Center for Instructional Innovation, has used Rails to develop tools for the teaching of critical reasoning skills. These tools have proven valuable in teaching undergraduate students and work is in progress to apply these tools toward the training of faculty and staff to correctly understand issues relating to the Americans with Disabilities Act as well as other institutional training needs. This talk will cover the path taken to fulfill this need as well as best practices and lessons learned in using Rails to develop applications in higher-education environments.

Full-stack Accessibility

4:25—5:15pm, Oregon Ballrom 202

At the University of Nebraska’s Accommodation Resource Center, Rails has been an important topic of interest. First, we use Rails as the platform for our web development, including our online presence, an expert system for making better student-technology matches called Accommodation Solutions Online, and a proposed social networking platform to match disabled high school students with post-secondary mentors. By extension, the topic of Rails accessibility is obviously of interest to us and we have been examining ways to make this process easier as well as watching what has been done by other developers. At the last RailsConf, there were a number of people who expressed interest in this and some advances have been made in the last year. This session will also look at what the concerns were and what has been done in the last year to improve this situation.

These are my first two non-local presentations, so I’ll try to make them interesting ;)

The last day of RailsConf 2006 and a few more sessions to remark on.

• I started with the session that Mike Pence gave on OpenLaszlo. I’ve kept up with this project for a little while now and am glad to see that the DHTML support has been released now. Unfortunately, accessibility is not a strength of this implementation and it seems that the Laszlo community is far too focused on client-side eye candy and not enough on the development of good applications. I might be misunderstanding the intent but, in general, it felt like there was far too much Laszlo prosletization and not enough demonstration about the OpenLaszlo on Rails project that I think most of us came to hear about.
• My second session on OpenID was a little more technical in nature than the first, largely because Matt Pelletier has actually worked with OpenID service implementation at EastMedia. Unfortunately, it seems even more obvious now that this technology is still very much under development and in need of design work. Still, I’m hopeful that this kind of technology will become more refined and present as we move forward.
• Next up: Stuart Holloway starts looking into MetaRails and diving into the rails sourcebase in order to build more powerful applications. Much of this is still over my head, but I already see a lot of value in learning about these underpinnings of Rails and, for that matter, of Ruby itself. Slides are available here.
• Had a really nice discussion with Dave Thomas just before lunch that was really informative about a number of the projects I’ve been working on. Dave is a great guy and, by my experience and others I’ve met here, incredibly generous with his time and advice in the middle of this conference. Thanks Dave!
• One last session—James Duncan Davidson has a nice collection of experiences and case studies to inform about deployment. Much like Ezra’s session yesterday, I don’t think I have a need to work with the scale he describes yet, but I think there are a number of things that I can apply to my projects in the near future. The land mines, on the other hand are really good and really important—and I’m glad that I’ve already addressed most of them!
• Core Panel Meeting Highlights
  • Django made a big point of removing magic. Ruby/Rails thrives on magic.
  • 1.1.3 most likely released today. 1.2 in the next couple of months with RESTful additions. 2.0 will involve removal of things that don’t belong in core and cleanup and will be released when appropriate.
  • ActiveResource will not be as rich as ActiveRecord—We’re not cloning WSDL.
  • Much new functionality will be in plugins rather than through the core to prevent J2EE style bloat.
  • Accessibility is something the core team is interested in being educated about. Preferably they will get this education through well written patches.
  • Fixtures need improvement. Creation and management costs need to be lowered—probably through the use of scopes
  • To get a patch noticed:
    • Build it with tests, documentation, explanations, and motivation.
    • If the implications are large, try a plugin first.
    • Don’t take non-application personally. Standards are high and the core are busy with projects.
  • While a non-profit foundation could probably be created quickly, there doesn’t seem to be a need for that at this point in time.
  • Armageddon (push AJAX)—not being worked on by DHH, but other people are working with the idea and trying to make something real of it.
  • Repairs of CGI.rb are not as important as at one point, but, if the patch is done and refined, there might be considerations, but this is probably better targeted at the Ruby team.
  • Call for help with documentation. People who don’t completely understand the framework are possibly the best to submit documentation patches.
• Closing: Don’t make this a once a year thing. Keep this dialogue moving. Things have already begun to move during the conference—for instance, Barnraisr has already been deployed to run with the ideas from Nathaniel Talbott’s talk last night. Take this drive and these challenges from RailsConf 2006 and run with them.

More impressions and experiences from the second day of RailsConf 2006

• Justin Gehtland had a talk about Rails and AJAX this morning that was promoted specifically in Dave Thomas’ keynote yesterday morning. The announcement of Streamlined was definitely worth the time and the rest of the talk was excellent as well. Slides are available here. Some of my favorite quotes:
  • DHTML and Javascript were cocaine—good but expensive. AJAX is crack… and Rails is the free vial.
  • There’s Javascript in my Ruby… Kind of like a peanut butter cup of EVIL.
• My next two sessions were on DSLs. Steven Hammond and Bill Katz both gave good presentations, but I felt the first was a little application light and the second explanation light. Perhaps this might have been better as an extended length team presentation? On a side note, it was nice to meet Glenn Vanderburg and chat for a few minutes about the non-technical.
• Ezra Zygmuntowicz had an excellent piece on Rails deployment. The scaling is something that none of my current projects need YET, but I’ll definitely be looking into Mongrel and BackgrounDRb.
• One of my highlights today was the session on accessible design given by Jason Kunesh. I don’t think that a large number of the audience were satisfied with the presentation, likely because of things that need to be said on this front are not necessarily comfortable ideas for content developers. This lead to a great offsite with myself, Jason, Peter Krantz from Valtech AB, Jeremy Seitz, and several others. In general, we agree that accessibility is one of the opinions that Rails should hold for its developers and hopefully Justin might be interested in accessibility implementation for Streamlined.
• Mike Clark was back for a discussion about testing. The session was great, with substantial advice for approaching the testing problem, and I’m interested in moving toward some integration testing in my own applications. The only downside for me is that much of the presentation borrowed heavily from the Depot application in Agile Web Development with Rails 2E.
• Nathaniel Talbott spoke about homesteading as it applies to software development. I can’t really do justice to the discussion, so hopefully the presentation shows up online sometime soon. Unfortunately, I heard a number of attendees muttering in the hallways that they were disappointed that he didn’t speak more directly about consulting and project development. On the contrary, Talbott’s presentation had everything to do with life, creativity, and pursuit of goals. If you couldn’t relate the things he said to your profession, you may want to carefully consider whether “Buying the Ticket” is really the right thing for you.
• I’m sure that there are already many good summaries of David Heinemeier Hansson’s keynote written, so I’ll just hit some highlights. After DHH started with a bit of (well deserved) ego display, he addressed many of the ideas from RailsConf thus far. He maintains that the opinionated method has made Rails what it is today and that this is not the time to start bending to the world. Possibly my favorite quote of the conference: “Indulging in insanity should hurt!”. He further explained his intent to make Rails more RESTful by refocusing on the CRUD. For a bit of excitement, DHH also introduced ActiveResource, which will be a framework for utilizing web services and mentioned the possibility of abstracting ActiveModel as a common base for ActiveResource and ActiveRecord—Interesting things ahead.

Just some general thoughts and impressions about the first day of RailsConf 2006

• Dave Thomas started things up with a list of challenges for Rails to rise to. Definitely a lot of things to agree or disagree with, but a lot of good context for the rest of the meeting. I’m now looking forward hearing Justin Gehtland talk about what’s been going on over at Relevance LLC.
• Introduction to Capistrano—not a bad session, and I think Mike Clark was a great speaker for this topic. The main snag hit when about half the audience seemed to be looking for a more advanced look into multiple machine deployment and the rest (myself included) were looking for a good general usage pattern for the tool.
• Jim Freeze’s Rails and Asterisk talk wasn’t particularly relevant to the things that I’m currently working on, but does provide a nice overview of a different usage pattern for Rails than I’m used to seeing. It’ll be nice to see if other usability enhancement solutions like this begin to surface.
• It doesn’t seem nearly ready for prime time to me yet, but Kiran Dandekar’s presentation on OpenID has me somewhat interested in the possibilities for this if some of the major goals can be achieved. Also had an interesting side conversation about identity issues and the consequences of the coming generation where “real” and “digital” aspects of the world don’t delineate as comfortably as in the past.
• In the realm of demos, there were definitely a number of interesting projects and it was good to see the directions people are taking Rails apps, but the show stopper was almost certainly the interface work done by FiveRuns. I’m definitely looking forward to seeing their beta.
• Martin Fowler gave an excellent keynote on what he considers to be the advantages of Rails, and Ruby in the broader sense. In summary:
  • Opinionated software is important. By experience, we know that building the “One True Solution” just doesn’t work.
  • Our paradigm allows for a stronger transition toward developing conversational software. This is Good for us and for our collaborators.
  • At the heart of Ruby and Rails is a community motivated by agility. This places people over tools and is a Good Thing.
• Paul Graham spoke about the “Power of the Marginal” reminding us that, by choosing to live at the margin, there are freedoms to take risk and innovate that are simply not accepted among “insiders”. Graham also tells the cautionary tale of how to be careful if you suddenly find yourself among the “eminent”. Some thoughts:
  • Great ideas come from the margins - but - everyone who has them looks down on them, including those who have them.
  • Don’t learn things from people who are bad at them.
  • Not having enough ridiculous ideas means that you’re thinking too conservatively.
  • The rise to eminence is like the yeast which is killed by the alcohol generated by its success.
• And Why the Lucky Stiff performed in the evening. I’ll have to admit, I left the show a little early, but I’m just not sure that Why is my thing.
• Spoke with DHH for a few minutes before the dinner break, and I’m impressed. He still seems to be in a bit of shock about the rapid growth of Rails but seems to be keeping himself very much accessible and real. I look forward to his keynote tomorrow night.
• Jay Zimmerman was also great to meet and speak with for a while. He’s been handling the challenges of pulling this conference off incredibly. As it turns out, 550+ may be taxing the abilities of the site a bit (I was in three sessions today that required several dozen more seats and ended with many people standing or on the floor—PragDave has photos) That combined with the presence of too many good sessions means I’ll probably be ordering the slides and video tomorrow.

RailsConf 2006, the first international rails conference has been announced for Chicago this June, and it looks like I'm going to get to go! I've been using Rails for about a year now, and it has made my software development life easier in ways that I can't begin to describe.

Update [2006-02-08]: RailsConf 2006 is sold out. Should be a great time—see you there!

I've had some time off, so I finally got around to bringing this site back up to speed. After a long search for new software, I finally decided to create a custom solution that I'm calling CommonPlace. This is a Rails application named after commonplace books, which I was first introduced to via Lemony Snicket's A Series of Unfortunate Events. If you've been to the old version of the site, you'll notice some content that wasn't there. I've been maintaining a “shadow” blog during the past months and decided to import those posts along with the archives from the old Semeiotike. Among other issues, this is why the dates on posts before this point look mostly contrived.