Djangocon 2012 First Day

The Caktus team arrived in DC for Djangocon on Monday evening, 9/3 and I've been keeping notes about what the presenters have been discussing and generally what the mood and experience have been like. I am taking a few minutes out of my Tuesday schedule to share with everyone how Djangocon 2012 is shaping up! I stayed in Track 1 throughout the day, so I can only report on talks given there.

Notes from Erik Sterling's Keynote speech:

  • Politics is a group decision making process, but "Politics, obviously is a dirty word"
  • Going to lunch is politics: where do you go? Who is going to join? Who will pay? Etc.
  • Political life now consists of disagreements about similar micro-political decisions.
  • But whose voices are going to be heard in the decision making process?
  • This is the question that underlies political legitimacy: if voices aren't heard, the process becomes less valid.
  • "Often the explanation of a political decision covers up the process, leaving out how we got there."
  • This exacerbates the feeling of legitimacy if the outcome isn't desirable to all parties.
  • People's interest in the outcome of group decisions drives them into politics.
  • What politics you are interested in reflects what you are passionate about.
  • If you are passionate about software legislation, for example, chances are you care about that.
  • There's always a choice when getting involved in politics: are you satisfied influencing the decision or do you need to take it over?
  • Student government, boards of education, congress all represent taking over the process
  • Question: how many people got a speeding ticket in the last 3 years?
  • Of those that got a ticket, who got one locally?
  • A very small number of people raised their hands.
  • How many people know who their local representative is?
  • Do we know their name, have we written to them?
  • A small number of people raised their hands.
  • Eric made the point that choosing town speed limits are a great example of outcomes over which we had a fair degree of control.
  • "There's no question that money plays a part in the making of decisions."
  • The fundamental contention in politics is between large players and the small players that outnumber them.
  • Very often the asymmetry of power comes down to information access.
  • "How can we design applications that level the political playing field to get information to the small players?"
  • Eric thought that we (here at Djangocon) were at the center of making that happen.
  • The big factor that enables large corporations to wield so much political power is that they have framed the political discussion such that ordinary people don't want to get involved.
  • Because it's dirty, unclean, unsavory, they control the process to a greater degree.
  • "There's a great deal of indirection in politics."
  • He described a gun control anecdote, which I suspect was largely lost on the audience.
  • He described how a scramble to ban ecstasy created a boom in drug sales.
  • Here in the US, power doesn't grow out of the barrel of a gun, "it is inconceivable that Americans could resist government action [with force]."
  • Information and knowledge are the key to political power here in the US, and people building software have their hands on the levers that control who gets what information.

Throughout the keynote, Twitter was pretty lively with people criticizing his keynote, and I felt that it did meander and wasn't terribly effective at conveying it's message. Still, the point is valid and bears reiterating.

Mark Lavin on Maintaining your sanity on an open source project:

Mark Lavin spoke about different methods of maintaining your project while maintaining your sanity. Highlights included using dynamic documentation, testing your models, and being supportive of contributions (to name just a few of his points). Several of the questions centered on when and how to get started publishing, Mark suggested that you should launch early: "if you're not embarrassed at v1.0, you probably didn't release early enough."

Mjumbe Poe: Creating Dynamic Applications with Django and Backbone.js

Started by explaining what Backbone.js is and how a person familiar with Django might think about it. Several of the distinctions included the differences between how central Events are in Backbone versus signals in Django and the distinction between routers and url patterns. The RESTful API was sidelined in favor of an example based on the Django Polls app. Mjumbe showed the polls app and introduced Hijax. The goal became removing the pageload when you submit an answer to the first question. Mjumbe relied on live examples, which proved to be slow, but he didn't run into any significant problems. The time did not permit going through the entire example, the talk concluded a good ways before the tutorial/example was complete. There were several questions at the end, but I wasn't able to get much out of them.


I got to meet up with some folks from SocialCode and CBS and was able to chat over a very healthy meal. Like last year, the meal was pretty loud, but I was able to meet some new folks and get a sense of how other people are using Django and Python. Everything in the Hyatt is pretty convenient, which is a definite plus.

Jacob Burch: A Gringo's Guide to Internationalization:

I missed most of this one because I was tending to some work, but my impression was that it was both a) well-organized and b) helpful. There are about a half dozen questions at the end, and I believe that the slides were put up for people to review. The only thing I was albe to take away: i18n stands for i + 18 internal letters in the word "internationalization" + the letter "n." I had always wondered about that!

Karen Tracey and Colin Copeland: Openblock: Overview & Initial Experience:

Colin and Karen did an excellent job of presenting about Openblock and the project that we have been working on based on the Openblock code, OpenRural. Topics of discussion included a brief history of the Openblock project, the application of the existing Openblock code to the OpenRural project, with a particular focus on some of the geocoding problems that we have stumbled upon and solutions to those problems. Since the talk ran long, there weren't many questions.

Russel Keith-Magee: DSF Keynote:

The talk opened with a basic explanation of what the DSF is, how things are going (Dan's resignation on his move from Media for Media, and Alex Gaynor joining the board). "We would like to expand this group (DSF membership)," he made clear that people can be part of the DSF membership without being a Django core committer. Corporate members are welcome, there are three levels of membership: small, medium, and large, ranging from $500-5000 per year. The corporate sponsors got a round of applause, and Russell went through the appropriate legal uses of the Django logo. He mentioned that they are looking for big ideas: what could the DSF do that would be audacious? What about a comprehensive Django book? Suggested that bug bounties might be something to consider. Essentially turning donations into bug fixes, that sounds interesting! Going a bit bigger: what about funded development? Summary: "The DSF is here to help make the Django community as awesome as it can be." He had a good closing line: 'Think big. Act big. We'll bring the checkbook."

I missed the API Design Tips by Daniel Lindsley and the Lightning talks, as I had to setup our table for the sponsor's dinner, which was more of cocktail hour. The attendance at the event was surprisingly high and people migrated from there to the Github Drinkup which was located about a ten minute walk from the hotel.


My impression of the first day was that it went very smoothly, the talks were well received, and with the exception of some political tension during the opening keynote, everyone is generally having a good time. I am looking forward to the 1:30PM talk by Julia Grace about bringing new devs up to speed on Django.

I took notes on Selena Deckelmann's excellent keynote about fixing computer science education and will hopefully post that tonight or early tomorrow morning.

New Call-to-action
blog comments powered by Disqus



You're already subscribed