Six weeks into our Urban League Heat Pump Accelerator project we have gotten into a rhythm. Our data analysis team finished an exploration of the data on housing in the neighborhoods that Urban League will reach out to. They are working with Urban League to develop the methodology to rank households for outreach. The app design team has created a prototype in Figma. They continue to add designs for new features like a mobile map view for the outreach workers. Finally the backend team I am working with has created the ability to sign up for user accounts and is working on the initial data import. With five to ten people per group we are making good progress!
One big challenge in working on a volunteer project like this is choosing which bottlenecks to take on and which to neglect. Unfortunately I cannot do everything. I have been spending time coordinating the backend team and assisting members with technical challenges so they can do the work of the project. This has meant I do not have the time to onboard new members. When new folks show up and have nobody to pay attention to them it is a bummer. However, attending to them would mean not making progress on the product. Ideally other people would step up and onboard new folks, but when most people sign-up to volunteer this is not how they want to contribute. Until one of the other members is comfortable or interested in orienting new folks, we just live with the fact that new volunteers will not have a great personalized experience because we have existing volunteers working hard on the project and pretty good retention.
As my backend engineering group has been working on the data we have been finding some funny things in it. The latest one was they were surprised to see errors when importing addresses because there were states longer than two letters. This does not make sense if you are in the United States, but we saw folks try to fit a round peg in a square hole: someone had a mailing address in Canada so the state was listed as Canada1. If the database designers in the City of Boston had anticipated this ahead of time, this would have been in a country field, but that does not exist in their application. I am sure this works fine for creating mailing labels, but is a little bit of an annoyance when doing data analysis.
CA would presumably have the letter land in California. ↩