Zumbatech takes on #hackforchange

On June 6, 2015, a team of engineers from Zumbatech decided to contribute in an all-day hackathon event called Hackforchange. This is a national effort for civic hacking that brings engineers and designers together to make a positive impact in our communities.

Let’s not re-invent the wheel

The day before the event, our team got together to figure out what we were going to work on that would give the most impact to our community. We chose to work on generating visualizations of Florida vendor transactions. The first thing we noticed is that a couple of projects had already been underway to create restful APIs and alternate data formats for this data. We decided that an API that is specific to this data set wouldn’t be very reusable for other data sets, and it would still take an engineer’s effort to visualize the data from those APIs.

We wanted to make something that is generic enough to work with any sort of data set, be flexible enough for other engineers to create tools and visualizations via an API, and be easy enough for non-engineers to construct visualizations that fit their needs. A daunting task, especially for it to be mostly completed in a single hackathon. After some planning and discussion, we came up with a solution ready for hacking!

Hackathon

Bright and early on that Saturday morning, we arrived in the LAB Miami offices to work on a project we called Datamnom. The idea of the project is to make a generic ingestion program that can take in multiple data sources and populate an Elasticsearch index. Once the data is in Elasticsearch, a tool called Kibana can be hooked up to the Elasticsearch index we populated to create visualizations.

After writing a prototype nodejs program and setting up a Vagrant environment, we had Kibana up and running with data to visualize:

Kibana running FL Vendor data

Here is team Zumba demoing our work to the Florida CFO, Jeff Atwater:

Team Zumba demoing Datamnom to Jeff Atwater and others

via Miami Herald

Presentation and Reception

After importing about 8 years worth of Florida Vendor transactions, the team presented our idea and applications to the group. Other groups that were working with this data set decided to use our tools to make their own visualizations.

Conclusion

We had a fun time at LAB Miami hacking together a project we think can really help lawmakers, researchers, and reporters visualize public data in a way that allows them to ask the right questions and help our communities.


« W3C Push API Crash Course

Comments

comments powered by Disqus