Over the past few months I have been involved with the development of uladoo.com. Uladoo is a small, in-house, collaborative web project between Atomic Object and David Christiansen. Uladoo provides a simple means of creating, updating and sharing charts for people who desire to track their daily activity. Uladoo uses Twitter for creating and updating charts.
David had the idea for Uladoo in the midst of a weight loss effort. He was already an active Twitter user and thought it would be convenient to be able to tweet his weight and have those tweets tracked.
I thought David’s idea was interesting because he identified that Twitter could be used as an application’s data source. At the time, most of the applications related to Twitter were on the surface of the Twittersphere, facing inward. Twitter clients, tweet management tools, and tweet analysis were all hot. These kind of applications were aimed at enhancing and organizing a user’s Twitter experience. David realized that we could think of Twitter as a communication protocol to power Uladoo. Users could input their chart data through a multitude of Twitter clients and we could use the Twitter API to pull our application data.
Because this project was in-house, we had the luxury of being able to exercise our product design and development practices as we saw fit. We viewed the idea as experimental and decided to forgo user research. We decided to release a minimal product that was easy to learn, account free, and simple to use. If the idea took off, we’d develop it further.
Being account free was a curse and a blessing. During our discovery and design phase we kept identifying all kinds of complexities. What happens if a user tweets to the same chart twice in a day? What happens when they skip a day? What level of granularity should we display the data? We kept getting bogged down into the details and started to think that a user would have to provide preferences. We didn’t want users setting preferences via Twitter because we thought that violated our principles of Uladoo being simple and easy to learn. Our principles forced us to make judgment calls and design the behaviour in accordance with our idea of the general case. Our self imposed constraints allowed us to move forward and not sputter out during the design phase.
We were able launch Uladoo within five weeks. We spent about a week in design and discovery and four weeks in construction. It was a great experience participating in a small, focused project. We’re continuing to add features to Uladoo. After we had about 100 people using Uladoo, we decided to provide a 10, 30 and 60 day view of user charts. We just added a feature allowing users to embed a chart in their blog or web page. Check out the Uladoo Blog for updates on new development.
