This weekend I'm participating in Startup Weekend DC, which is, to quote the blog - "an idea, an experiment, a chance to gather the tech community and create a company over one jam packed weekend." I just returned from the first day.

Choosing An Idea

Our first goal was for the day was to choose the idea for our startup. After about 3 hours of several rounds of voting and group discussions, we settled on Social Networking for Condos. The idea was to build a social networking application that condo associations could use to help foster communications between residents. I can't say I was too jazzed about this idea, since I think the whole social networking thing is overdone. That said, it is a good application for the large team of developers we have (30+), since there are definite, concrete sections of functionality that can be broken down and divided among teams.

After making the decision, we broke into teams by area of expertise (Development, Creative/Design, User Experience, and Marketing/Business Development.) In my group - the developers - we quickly decided to use Ruby on Rails since a majority of people were familiar. After finding out that a box was being provided for development, we also decided on using Apache/Mongrel/MySQL as the rest of the stack, and subversion for source control.

Defining the Project

In theory, each group was supposed to focus on their niche and then come back together with something to present to the other groups. This is where things started to break down. Each different group had their own idea of what the site was going to be, who the target customer would be, revenue streams, etc. After a great deal of debate between the groups, a smaller group of us from each team met to come to agreement on the most very basic functionality. After even more debate among the smaller group, we could all agree that we were building a social networking app for condos that would help solve condo-related communication problems.

This pretty basic idea was presented back to the whole group. More discussion followed about specific features, and people were still pulling in several different directions. The option to completely scrap the idea and start with something new was presented, voted on, and scrapped. It seemed to me that the organizer was getting very frustrated at our lack of progress at this point. It was decided that the small group would barricade ourselves back into the room until we came up with a vision for the product.

Yet more discussion and debate followed, and we eventually settled on some base pieces of functionality that we could all (for the most part) agree upon. Two people volunteered to stay and put the thoughts and ideas together into a short PowerPoint to present Saturday morning, and the rest of us packed up to head home.

Overall thoughts for the first day:

• Though I wasn't really excited about the Condo Social Networking idea, I quickly started to see how such a product could be useful, and potentially marketable.
• I was surprised how many directions the different groups took the idea of a social networking app. Social networking apps are pretty well understood at this point - I didn't think it would be difficult to all agree on at least a base level of functionality.
• Sometime during one of the endless rounds of discussion someone said that people needed to make sure they were arguing to attain consensus, rather than arguing to win the argument. He said arguing to win an argument seems really prevalent in DC - can't say I disagree with that!
• I'm a bit worried about the size of the development team. I think, even with using a pair-programming approach, we're going to be stepping on each others' toes quite frequently.
• That said, I can't wait to get done with this idea-generation phase and actually get into some coding!
• I'm hoping to mainly fill a testing role, and getting to use some of the rspec fu that I learned at the Pragmatic Studio last week.