We made this map for a couple reasons. First of all, it’s fun to watch the buses move. And it’s an interesting picture of the overall status of the system: you can see the city in motion. There’s a couple other uses; you can zoom in to your house and use it to know when to head to the bus stop (though we recommend using an app like Transit or Stop313 for that).
Here’s how we did it: The data comes from the TextMyBus API. You may know TextMyBus from it’s cell phone fame: you can text in your location to 50464 to get real-time bus arrival info. Over 42,000 people have sent & received more than 2,500,000 messages. But the data behind it is also free & open, and it’s used by several mobile apps. We’re using it here to make this little map, and we’re excited to see what others make with it.
There are some caveats to the data. The data isn’t accurate to the second: buses only update their location every minute or so, and it takes some time to receive and process the data. The map also doesn’t show untracked buses. That means buses that are scheduled to be on the road but aren’t won’t show up. It also means that if the bus’s GPS is broken, we won’t display it. If you notice anything wrong (like a bus not coming when it’s shown on the map), leave a comment here or email us at firstname.lastname@example.org.
“This map aims to show the current air quality in Detroit’s public schools, investigating whether more school children are exposed to oxygen from parks or carbon dioxide (and other pollutants) from surrounding industries. I assumed that larger acreage/area translates to more a more intensive use of either park/industry. Schools close to industries are highlighted in navy, while schools close to parks are highlighted in dark-green. A comparison of the frequency and size of these two categories illustrate the conditions in which students study in.”
There are several major problems with this map that make this a good place to discuss map and data design:
Use meaningful measures. “O2 and CO2 coverage” are not units of measurement. Outside carbon dioxide levels is not necessarily a relevant measure. Particulate matter is often used as a measure of air pollution.
Explain the analysis. It appears the coverage is mapped by simply drawing buffers around parcels, not by any scientific measure. Parks don’t produce plumes of oxygen; industrial areas don’t produce plumes of carbon dioxide. It’s unclear how the size of the buffer was set.
Validate your assumptions. The authors assume that parcel size corresponds to intensity of use. Absent other evidence, there’s no indication that is accurate. For example, the Packard Plant, which is zoned industrial, has been unused for decades. It probably does produce a lot of pollution from scrappers’ fires, though, maybe even more so than larger nearby active industrial areas, which are often regulated.
Make sure colors are relevant and high-contrast. It is hard to distinguish schools near industry (dark blue) from industrial parcels (shades of purple).
Hamtramck and Highland Park are uncovered on the map. (We should start a library of maps missing the middle of Detroit).
A map may not be the best way to show the data. The authors write, “A comparison of the frequency and size of these two categories illustrate the conditions in which students study in.” A better description might be a chart or table that lists student population near parks, near industry, or both. This would be more actionable — interested parties can focus on the locations that have the most students. Adding grades might help even more by identifying younger students who might be at more risk.
Cite your sources. The data sources are not listed here. Detroit Public Schools has closed or transitioned dozens of schools in the past decade, and we can’t tell from the map how old the data is.
We’ve seen quite a few maps of community resources in Detroit. But where does the data come from, and how do we keep it fresh? Most projects rely on static lists and maps that don’t change with time. Here’s a new approach.
That means we can fix or add anything on the map, from roads to points of interest. Hundreds of companies and organizations around the world use this data. Take a look at this beautiful report to see the breadth and depth of this data.
We need your help. If you click the “Supermarkets” tab, you’ll see why: none of Detroit’s supermarkets are on the map. By using a public resource, instead of a closed list, we stand a better chance of keeping the data accurate, up to date, and available for everyone.
Join us for a DetroitWiki + OpenStreetMap editing party at D:hive on Wednesday!
We’ll demo editing DetroitWiki, a collection of Detroit-related info that everyone can contribute to. We’ll also teach you how to edit OpenStreetMap, the map of the world that’s used by people from small community groups to big companies like Foursquare and Nokia. Make sure your neighborhood is on the map!
If you’re already a veteran editor, join us in contributing on your favorite site (or both!), and help get new editors on board.
We’ll provide tasty food and drink — all you need is a laptop. (let us know if you don’t have one, and we’ll try to dig some up)