More people would likely ride the bus if service were more convenient. But would more people use the existing service if it were just easier to understand?
A few weeks ago, this blog discussed “spider maps,” used widely in London. They are one tool to better communicate bus service. I created a similar map for the H Street neighborhood:
Rather than attempt to communicate the entire route network, a spider map only shows the routes that serve stops within walking distance. Like London’s famous Tube map, it also forgoes the geographic accuracy of a street map for a simplified diagram of connections and destinations. It answers the questions “where can I go from here?” and “what bus do I take to get there?” without adding unnecessary information.
When I lived in South London, with no car and a long walk from the Tube, I moved across the city by bus using spider maps. They gave me a clear mental image of the destinations I could travel to from my regular neighborhoods. I also knew that if I found myself someplace new, I could use these maps to easily find a bus home without pre-planning my trip.
After moving to DC a year ago, still with no car, I’ve yet to figure out where a bus can take me from my own neighborhood. For many trips, I’d be happy to take a bus, but because of the effort it takes to figure out the system, it’s usually easier not to bother.
The Mobility Lab is hoping to develop a program to generate user-customized spider maps, and will work on it at Saturday’s hack day. In anticipation of this project, and to demystify one corner of the bus network, I’ve designed this modified version of a London spider map for the H Street NE area, designed to fit in poster slots at bus shelters.
Consider your challenges if you’re a new bus rider in this neighborhood. Some stops have no information at all beyond the route number. If you’re not already a regular on this route, you’re out of luck. Many stops have a small route map on the sign post:
There’s some useful street information in there if you can find it, but it’s far from ideal. How do you know where you are this map? If this bus doesn’t take you where you want to go, is there a nearby bus that will?
You might try to get a bigger picture with the Metrobus service map available online and posted at some bus shelters. It shows all services on a street map of the entire District, which is much more information than you need, and it’s quite difficult to decipher. Here’s the H Street area:
To figure out which of those red lines connect to which other red lines, you’ll have to try to connect the dots between labels, and in many cases it’s just a guess. The Metrorail red line looks like a bus line. The B2 is green because it goes to Maryland, which is helpful if that’s where you’re headed, but doesn’t really matter to you if your destination is on the other 95% of the route. Downtown is removed in a separate inset. Figuring out how you can get across town will require patience and determination. (The older version of the map actually used more colors, but Metro changed to this more confusing version in 2009.)
A few shelters offer an improvement on this map that highlights only the routes from that particular stop. This is the right idea, and it’s an easy way to quickly improve legibility. But it still simultaneously gives the H Street rider too much information (a street map of Tenleytown, for example) and too little.
Some maps would be more useful if they showed additional routes nearby (top left). Others (bottom) show two services equally, even though one only runs on weekday mornings. At a stop on Bladensburg Road, the map (top right) says that the X2 down Benning Road stops there, but it’s actually across the street.
By this point, you’ve probably given up and are taking a cab. It’s a shame when transit is under-utilized because of poor information.
The spider map makes bus service more visible and understandable by focusing on the only the information relevant to your current options. Its focus area is more than a single stop, but less than the entire system. Within the focus area, it shows clearly where each stop is and which line it serves:
Outside the neighborhood, each separate route has a separate line. The diagram gives the names of all major streets and neighborhoods these routes serve, but it doesn’t add clutter with a full street map. It shows all Metrorail connections, but not the bus routes outside the focus area.
My design makes a few modifications to London’s in adapting the style to DC For example, in London, you can expect most buses to run with reasonable frequency throughout normal operating hours. This is far from true in Washington. If you see the X3 on a map and think it will be a convenient trip from the Atlas District to U Street, it will be very important for you to figure out that you can only make that trip on weekday mornings. My design gives limited service routes a different graphical treatment and a clear label.
I also include a table of approximate service frequencies. (Riders would ideally check their route’s detailed timetable at that route’s particular stop.) Metro’s current stop-specific maps offer this table, but not for the nearby services that may provide better options.
In addition, London’s maps generally cover a smaller area, perhaps a major intersection or the roads surrounding a rail station. I could have created two separate maps for this area (one around 8th Street and one for the Starburst intersection at the eastern end of H Street), but these intersections have fewer routes than the equivalent in London. In Washington, there’s a greater likelihood that you’ll have to walk farther to get the bus you need, so I use an expanded area.
This map shows the bus routes through the central street map, whereas London omits the route lines and directs riders to the appropriate stop using a system of letters. In part, this is to avoid the spaghetti mess that would result in showing routes through intersections like these:
It’s also necessary because in London, adjacent stops may serve different routes on the same street, and the letter system helps riders identify the stop they need. It’s true that the exact routing is unimportant; all the rider needs to know is where the stop is, and where the bus will go. But the DC routes are straightforward enough to show, and with only a few exceptions, every bus serves every stop it passes. (The exceptions are made clear with symbols on each individual line.)
Finally, while the spider map style favors graphic economy over geographic accuracy, I’ve included more geographic clues than London maps do. These include the rivers, the Mall, the District boundary, and the quadrants. London’s labyrinthine streets are famously difficult to navigate or conceptualize, but the L’Enfant grid is a coherent orientation tool for most Washingtonians. I take advantage of those mental reference points by maintaining some diagonal angles, showing most major turns, and placing all stops in their correct quadrant. This level of fidelity means the map doesn’t get quite the spatial compression of London maps, but it still saves space and removes noise from the full-scale District map.
One modification worth considering is adding information on bus transfers. London buses offer better point-to-point service, but in Washington you may want to know what other routes you can pick up. For simplicity’s sake, I’ve omitted them here, but a better design decision would be based on data on the frequency of bus-to-bus transfers.
The spider map is not a standalone solution. It works best with complementary signage, sign-post maps, and timetables that are stop- or route-specific. Personalized tools and mobile applications are critical rider resources as well, and the Mobility Lab’s projects hope to add real value here. My bet is that a system of spider maps—not cheap to create, but cheaper than expanded service or capital improvement projects—would increase ridership simply by making the bus less of a mystery.