The past weekend was my first Startup Weekend. I was somewhat skeptical if we could give build something good in a weekend. Nevertheless, I was still very excited.
My main goal was to deliver a working app by Sunday night. All the actions and decisions I made the whole weekend was influenced mainly by this goal.
Friday Night: The Pitches
The event officially started at 7pm on Friday night. I came to the venue about 30 minutes early and were able to meet a few of the attendees. Andrew Hyde and Tyler Willis welcomed the attendees and talked about the event.
Then came the pitches!
There were some pitches that caught my attention. Among them were the Ponzi site and Yelp for EULAs. I was going to pitch a Twitter app but decided not to. I just thought that it would be better for me join a team rather than convince other people to join me.
I was glad that the VCs gave feedback to the pitches. Dave McClure played Simon Cowell that night and gave an honest assessment of the pitches. Dave told us to present the problem first and then the solution. It was effective and everybody kept that in mind come demo night. It was refreshing to hear Dave's comments. As an aspiring web entrepreneur, I have talked to a number of startup founders. I have met Y-Combinator founders through my roommate and also other founders through tech events. We usually play nice to each other when somebody pitches an idea because we don't want to put a fellow entrepreneur down.
VC Panel. Image Courtesy of kyeung808.
After the pitches ended, it was time to form teams. I felt like I was in American Idol where contestants form groups for their musical number. I'm not a social guy so I dread situations like this. Fortunately, the ratio of developers to non-developers favored me. As a developer, I had the luxury of choosing teams I want to join in.
With this advantage, I set three criteria for my ideal app. They are the following:
- A working app can be developed over the weekend.
- The app is not based on a business built before the event.
- The app doesn't need a lot of content to become usable.
Using this criteria, I talked to the people who have made pitches. After half an hour, I didn't find one that satisfies all of them. I was hanging around by the pantry and stumbled to a bunch of guys who either didn't have a team or doesn't have a real team. The idea that was thrown around was an app that is like bit.ly but with analytics for different social networks. I decided to join the group and we discussed the idea. I wasn't really too keen on the app because it's not something I would use. However, I decided to stick with them because I felt that they are a bunch of smart and level-headed people.
Our team had 2 front-end (Amin and Eric) and 3 backend developers (Vaughn, Alex and me) plus 2 business guys (Matt and Anish). It was 11pm and the venue was closing so we decided to head to Mel's diner. we brainstormed a little bit more and went home afterwards.
Saturday AM: Evolution of an Idea
The next day, we met up first thing in the morning to discuss the app. I believed that the app needed some tweaking and I was glad that Eric stepped up and suggested we change it. He suggested to have a service that gives a short url for multiple links. I told him that several services with this functionality exist already.
This is when I started to really enjoy this event. The idea started evolving before our very eyes. Everybody made their own suggestions to make this app different. All had strong opinions but nobody was being overly aggressive in pushing their own idea. Consensus was finally reached. The app will aggregate multiple links into 1 short url and present them as tabs in an iframe at the top of the page.
Then came the technological decisions. The developers all have different skill sets. Alex is a Ruby guy, Vaughn is .Net and I'm .Net and Python/Django. I suggested that we use the Google App Engine (GAE) platform and they agreed. I was glad about this decision since I love coding in Python. Vaughn and Alex, however, were out of their comfort zone and had to learn Python and GAE fast.
We got down to coding right away. We set up an SVN repository and our development environments. I was impressed at how Alex and Vaughn picked up Python, Django's templating engine, and GAE easily. I sat between the two of them while we were coding and answered some questions they have. I have been coding in Python for less than a year so I'm not really an expert. Fortunately, James from img.gr was sitting next to us and lent his expertise.
At this time, we still did not have a name for the app. I realized that we had to come up with one soon because we need to get a domain name and have it resolved to Google by demo time. We met up and suggested names for app. We were hammering domain web sites to determine if the domain we want is available. Since our app is shortening urls, we had to get a domain name with few letters as possible. We looked at domains that end in 'ph', 'it', and 'me'. Again, I really enjoyed this part. We agreed and disagreed on different names but we eventually picked one that we all stood by with. Our app would be called hubb.me.
Saturday PM: Building the app
From mid-afternoon onwards was building the UI, coding and working on the presentation. We had issues merging our individual code changes but eventually figured out how to split our code. I also had a lot of hiccups trying to get our domain set up on Google Apps. Documentation is not one of Google's strength. Their articles on Google App Engine and Google Friend Connect are just insufficient. Their documentation on setting up your domain on Google Apps is no exception. You cannot find Google's nameservers anywhere on their site.
We called it a night by 11pm. I was pretty tired and I was glad that I was not the only developer on our team. I felt a little bad though since I heard that other teams didn't have any backend developers while our team had three.
*Part 2 (Crunchtime and Demo Night) coming tomorrow.*