Sunday morning I woke up first, a fair bit before we were expected to return. I separated the tests into an online and offline component and set up autotest. That put me in a good position to unify the API based models with the independently developed custom content models.
All my development thus far had been completely backend. Later in the morning we integrated the first draft of the actual app layout and set it up to return the results fragment.
The afternoon had a small interruption – Brad Keywell was supposed to give a speech before the presentations, but his scheduled got messed up and he came early. On the upside, it pushed back the stop-work deadline. On the downside, I was too busy working to pay very much attention. Of the few things I noticed: Brad says Ideas Week was his idea; ideas create business which improve the world. Be different – even though society wants you to be the same. His big trend is social networking plus collective action. They work by trying to convince themselves that their ideas suck – fail fast and only the strong survive.
In the afternoon Mingwei returned from a forgotten study session. Unfortunately, we were all heads down working on things, and had enough that we didn’t expect to be able to use any more API research. Missing time on Startup Weekend is dangerous because you can be left behind. Though I’ve never missed time, I’ve been in his shoes several times. I believe he was able to help out the business team a bit. He was also asking me about Ruby, so I think the brief exposure on Saturday may have proved infectious.
I also kind of wondered if some kind of SCRUM would be useful to keep people in sync and identify problems early. Another interesting aside is that Saturday the organizers came around to ask for our one-line pitch, and nobody was comfortable giving one, since we’d been bouncing lots of ideas around and working in separate teams. And organizing people was still hard – there was a little disconnect about what people wanted and what we were actually doing – it’s all very well to say “yeah, that sounds goods”, but that doesn’t mean I’ve finished the last task, or the next, and actually done it. At one point I had to say “If it’s less important than the last thing you asked me, put it on the todo wall.”
Team Pacha didn’t focus on a great presentation, or on much customer validation beyond a survey. The judges were looking for customer validation and spit-and-bailing-wire product tests. (I’ll write more on the pitches later.)
We did make Pacha something that looked pretty good, and returned some results – most of them canned for a few sample locations, plus some data from the easy to use WikiLocation API. I dropped Yelp due to the tiny API limit – we didn’t have time to test how the program would react to it running out.
We are going to talk about continuing. On Startup Weekend, everyone had made time to be there; now we have to deal with sets of very different schedules in order to get together again. Organizing people is still hard.