Visualizing Tango Tablet Data with Mozilla A-Frame VR Platform
This is a experimental visualization built from data recorded during a guided tour using the Tango AR Tablet.
The blue squares represent the tango tablet, it's location within the environment, and the direction it's facing. This is time-series data captured at half-second intervals.
The white walls represent the layout of the environment, minus any furniture. While the tablet is drawn to scale, the walls are taller than scale.
The 3D model is rendered in the Firefox web browser and completely navigable.
The visualization uses Mozilla's A-Frame VR platform.
Click here to view the 3D model in your web browser.
Base Flood Elevation Mapping
This web application finds the nearest flood elevation marker in the state of Illinois using data released by the Department of Homeland Security.
The geospatial requests are made through a web front end, to a Node.JS server, and a Mongo database. The system is hosted on Heroku.
This is the backend prototype for an Augmented Reality front end built using Unity 3D and working on Android and iOS.
You can view the web portal by clicking here.
Visualizing the Frequency of Ethanol Spills
Using government data tracking chemical spills, the investigative reporting team looked into the increasing frequency of ethanol spills taking place throughout the United States in the past 20 years.
This visualization was challenging to optimize and required reducing the data payload as much as possible in order to have it load more quickly.
Admittedly the graphic doesn't load or update as quickly as I'd like, but it was a great exercise in optimizations and opportunity to use new tools to showcase the data in a more enriching and compelling way.
Visualizing WTO Exports for Proposed Trans-Pacific Partnership (TPP)
This is an interactive graphic showing U.S. agricultural exports and imports with TPP trading partners.
It was built using Leaflet.js and d3.js, and is hosted on GitHub.
It is available here.
Property Record Search Database
Using the property records used through the investigative piece on homeowner’s exemptions, The Midwest Center for Investigative Reporting sponsored the build for a searchable database to accompany the story.
This application is built using Python and GeoDjango. Records and map points are served up from a PostgreSQL database and presented using leaflet.js. Search queries are handled by Elastic Search. The application is hosted on Heroku.com.
Click here to use the database.
Uncovering Multiple Homeowner Exemptions Claims
This project has been almost entirely a database driven story using off the shelf products such as PostgreSQL and SQL queries, with the presentation produced using CartoDB and some web markup.
Using property records for Champaign County, this story looks at how a loosely written state law allows property owners to claim more than one property as owner occupied.
My contribution to the project began with a simple database query on my own home, in which I noticed that the landlord was claiming a homeowner’s exemption on the house.
The curiosity of this situation led to further inquiries, and the investigative team found that the landlords are all behaving within the law, though the law as it is written is left up to the interpretation of the county supervisor of assessments.
The team further found that the tax assessor was interpreting the law differently then every other county in the state and in a way that generously favored landlords, leading to millions in lost local tax revenue.