Desert Code Camp Follow-up II
May 16th, 2010
Some random thoughts on the recent DCC:
About when I was going home it occurred to me that, in the entire day, I only saw one actual full-on presentation (Naum’s JQuery UI session). What’s interesting is that it didn’t occur to me sooner, because I was having a good time.
I gave a session at 9am, then saw Naum’s thing, then wanted to catch the MongoDB session, but the speaker never showed. However, the assembled did some self-organizing and we had lightning talks instead. It was great.
Then it was lunch, and then I went to learn about Node.js, only to find that the speaker was some person who was supposed to talk about MongoDB, and still MIA. This time I gave an impromptu show-and-tell of my Lilypad Arduino + Bluetooth + Android setup (though pretty much everyone had left the room by then to go watch laser stuff).
And then it was time for my next session, an intro to Ruby. When that ended it was just about the end of the day. None of the remaining presentations in that final hour looked compelling, I felt like I had a full day, and headed out.
I then discovered, as I was sorting out crap in my car preparing for the drive home, that I left my coffee thermos in one of the class rooms. Today I bought some Styrofoam cups and caps to keep in the car so that I can avoid ever taking the travel thermos out of the vehicle (except back into my house). BTW, the themos is quite slick, and the only one I’ve seen that is both spill-resistant and can be easily operated with one hand. Luckily, once I found a place that carried them, I bought a few, so I had two in kitchen cupboard. Lessons learned: redundancy is good, and plan for failure.
Another interesting observation: I had more people for the Haskell intro than for the Ruby session. The difference was not great, maybe a few more in the Haskell group, but still interesting. Feeling I get from some casual conversation is that people are increasingly getting hip to the power of pure functional programming and the pitfalls of OOP. They may not go learn Haskell, but they’re looking at it, and Scala and Clojure as well.
During the ride home I was thinking about the DCC format. I had a great time, but still was a bit bummed that I could not see David Koontz talk on Unity, or Remi talk about Chrome extensions. The schedule is more wide than deep. That is, there may be 10 things going on at once, while the number of time slots is no more than six. I would like to see shorter sessions times (40 minutes perhaps, maybe less), and fewer “tracks”, but have the day run until 6pm. You would then have fewer possible conflicts for your schedule, and get to see more sessions overall. I’d also like to see time set aside for lightning talks, too.
2 Responses to “Desert Code Camp Follow-up II”
Have some thoughts on this? See here for how to reach me.
May 16th, 2010 at 09:30 PM
What was disappointing for me was that a couple sessions I was interested in, the speaker declared that he had nothing prepared and was just winging it. And, funny thing, I remember the same exact speaker did the same thing last DCC. I understand that you don’t know what audience expectations and expertise level, but gosh, at least have a plan.
Don’t know that universally shortening the sessions is a good idea, but the tracks should be better planned (i.e., a couple for Web App development, one for MS platform, one for Apple Obj C/iPhone/iPad, one for DB, one for functional programming, etc.…). Maybe one could be standard, with longer periods for special hack sessions and some shorter ones for lightning talks or group discussions.
May 17th, 2010 at 11:50 AM
There was a fair amount of winging it in my sessions. For the Haskell intro I had spent a lot of time preparing possible examples, noting what to cover, and thinning out the interesting-but-too-complex stuff. But I didn’t have any real slides. I figured I wanted to show code, and use that for stepping-off points for assorted topics of interest.
I think having a flexible game plan gives me more room to adapt to the particulars of the actual audience.
For my Ruby session I had really nothing planned per se, but I’ve given other intro Ruby talks and knew what I wanted to cover and the main points to drive home. Other than my complete failure at attaching a video cable I think it went well. But winging it at the time of a talk still requires a good deal of pre-sessions thinking, and you need to have some considered options for what to say and where to go.
The tricky thing about planning tracks is that talk submissions are uneven. Where would Haskell fit? Hell, seems like all the Ruby stuff got labeled as “functional programming” :)
My main concerns are that most speakers do not make good use of their time, so shorter would work just as well for them. And the number of concurrent presentations means you missing way more than you are seeing. Out of 55 sessions, the most you could have attended was six. I suppose from one point of view that makes for 6 sessions you would not have seen were it not for DCC. On the other hand, I think there’s a way to adjust the schedule to get that number up to 8 or 10, which works out better for both for speakers and regular attendees.