Transit
Would personalized "spider maps" be useful for DC?
London has a complex network of train and bus lines, and a unique type of diagram to help people navigate them, known colloquially as the "spider map." As part of the work at the Mobility Lab, we're looking into generating these for the DC region, but going one step farther, and creating personalized ones for any location.
"Spider map" from Victoria Station. Click to enlarge (PDF). Image from Transport for London.
At each Tube station, a diagram like this shows each of the bus lines that travels from the immediate area. In the center, a geographically accurate map shows the nearby streets and bus stops. But instead of continuing the lines on a geographic map, which would either have to be very large or zoomed very far out, a spider map shows each line in the simplified format of the Tube diagram.
Someone trying to find a bus can therefore more easily grasp at a glance the location of that bus's nearest stop and what neighborhoods it travels through. It's easy to see all of the buses and their destinations, where they merge or cross, and more.
The Washington region also has a very complex bus network. The WMATA bus map is virtually unusable, partly because many lines are all the same color, but also because it's just so intricate:
Downtown section of the DC bus map. Click for full map (PDF). Image from WMATA.
But any individual person doesn't use all of the bus lines. You only likely use bus lines that travel near your home or work, or both. At any given time, if you're interested in a bus, you only wants the lines that stop near your current location.
Would it be useful to have a spider map in this case? You could get a spider map centered around your home or work. It could help you understand your bus options; many residents even with good bus access often aren't aware of bus lines that could serve their needs.
Transport for London's maps are hand-generated by a contractor, but we wouldn't do that. Instead, as one of the upcoming Mobility Lab projects, we're considering writing software to automatically generate maps such as these. They might be a little less perfectly refined than one made by a human, but could center around any point instead of just a small set (Tube stations, in TfL's case).
Or, are spider maps the wrong way to show bus information? Kerwin Datu criticized these maps, specifically their usefulness for tourists. Jarrett Walker thinks a "frequent network map" would be better (WMATA planners created one for our region).
Walker, though, actually doesn't end up recommending doing away with spider maps. He actually just thinks spider maps shouldn't be the only way to show the buses. He'd like to have both a frequent network map and a bus map that shows buses just for a few kilometers, the range where people are most likely to take a bus for non-commuting purposes. That map should also, Walker suggests, emphasize the most frequent services over the occasional or rush hour-only ones.
Using Walker's suggestions to create a Washington personalized spider map, we could imagine a circle showing all roads within a 10-minute walk of the center point, and then lines showing all buses and where they go for a few miles. Frequent buses could get thicker lines than less frequent buses, and rush hour buses lighter lines than all-day buses.
Major attractions could get prominent placement. For example, what bus goes to Adams Morgan, or Georgetown, or H Street? Many people don't know; such a map could make this clear, including which bus is most frequent and has the best span of service.
Not every map needs to serve all purposes. The standard Metrorail map is great for helping people figure out how to get from one station to another, but it is less effective at helping people know which station to use to reach a certain landmark, for example.
The purpose of this map project would be to help demystify the bus, showing each person what they need to know to understand and make use of their bus options. This should be a map that hotels could print out and hand out at the desk, or realtors could give to new home buyers. What would make such a map most useful?
 | David Alpert is the Founder and Editor-in-Chief of Greater Greater Washington and Greater Greater Education. He worked as a Product Manager for Google for six years and has lived in the Boston, San Francisco, and New York metro areas in addition to Washington, DC. He loves the area which is, in many ways, greater than those others, and wants to see it become even greater.   |
Comments
- Young kids try to assault me while biking
- Metro bag searches aren't always optional
- Focus transportation on downtown or neighborhoods?
- Endless zoning update delay hurts homeowners
- Redeveloping McMillan is the only way to save it
- DDOT agrees to repave 15th Street cycle track
- Vienna Metro town center won't have a town center
DC is such an easy city to spatially orient oneself. As long as maps include a bit of geography (Rock Creek Park, Potomac, Anacostia, major landmarks) for a touch of orientation: I'd be a happy lad.
I think the bigger limitation is with bus agencies having the resources to make tailored maps for so many locations... but I also often think that transit agencies too-rarely tap the power of the transportation geek community.
by Bossi on Aug 18, 2011 12:57 pm • link • report
Walker's argument with regard to frequent networks and frequency maps isn't just that we should highlight those services, but that those services should be structured in a logical way.
The diagram solution doesn't really address the fundamental problem - if the route structure for these bus lines themselves were much simpler, then it would be easier to convey that information.
by Alex B. on Aug 18, 2011 1:01 pm • link • report
Well then,let's just eliminate all bus lines except for the Q2. That will certainly simplify the system!
The "underlying problem" is not really a "problem" at all. It is an emergent property of a robust transit net.
by Dave J on Aug 18, 2011 1:08 pm • link • report
Actually, I do think our network is overly complex. Did you, for instance, that we have one route that runs one trip per week? Not even a round-trip, just a one-way trip. And it's not even a school route.
It's route A11. And it runs the route of the A12 for the first trip on Saturday. But because the Blue Line hasn't opened yet when it gets to Addison Road, it runs all the way downtown (to Federal Triangle).
But that line appears on the bus map (and on bus stop signs). So perhaps someone wanting to go from Federal Triangle out MLK Highway on a Monday afternoon waits, and waits, and waits, and never sees a bus.
Or consider the G13. It's basically the G14, except that on weekdays before 9:00am, it doesn't loop through the Beltway Plaza parking lot. It saves commuters headed from Lanham and Greenbelt to the Metro 3 minutes or so, and since the mall's not open, no harm's done.
But wouldn't it be easier to call it the G14 and then indicate in the schedule that it just doesn't serve that stop before 9:00am? I once encountered a commuter who insisted in the PM rush on waiting for the G13 because it was faster (by 3 minutes) than the G14. I tried to explain that the G13 doesn't exist after 9:00am, but she either did not understand or did not believe me.
Buses are complicated. And the map would be a lot easier to draw if we rationalized the system. But that's politically and logistically difficult.
by Matt Johnson on Aug 18, 2011 1:18 pm • link • report
I don't know if you've delved into the wayfinding literature but I highly recommend it. The Berger and Gibson books definitely.
by Richard Layman on Aug 18, 2011 1:36 pm • link • report
by Frank on Aug 18, 2011 1:43 pm • link • report
I think these maps would need to indicate major landmarks, neighborhoods, major roads and connecting bus routes and rail stations. I could go either way on geographical accuracy.
As for representing different spans of service, I think it is essential to deemphasize routes that don't run all day or don't run frequently. Or simply not show them. I'm tempted to suggest a 10- or 12-minute limit on these. Or have the user be able to choose when accessing the website what frequency and span of service to limit the map to.
As for rationalizing the bus system, it's definitely overdue. Once a system reaches a certain level of complexity, no amount of information design can make it easy for a user to quickly understand all the information available. You end up making maps that don't portray the full depth of the system's complexity, or a map that no one can understand. There comes a time when the underlying system should be simplified and rationalized. We're pretty close to that time with Metrobus.
by MDE on Aug 18, 2011 1:46 pm • link • report
Personally, I don't think the WMATA map is all that horrible. Before "real time," I would look at the map and figure out where I needed to go (or key it online) and I've really never had much of an issue.
The current bus schedules let you know which line serves a specific leg of the trip at specific times of the day.
by HogWash on Aug 18, 2011 1:47 pm • link • report
A couple points about the London maps. There is actually a spider map for everywhere in the city that’s served by buses, not just from tube stations. There’s no need to custom-generate a personalized map, because there already is one for your neighbourhood at the bus stop (and online).
Most spider maps show three scales. At the center is a geographically accurate map of the immediate area and its dozen or so bus stops. Landmarks are clearly shown and it’s suitable for navigation as a pedestrian. Outside of this is a zone in pale yellow naming every bus stop on routes in every direction for about ten minutes of travel time, covering your short-distance trips in reasonable detail. (Your Victoria example for some reason doesn’t have this.) Then the outermost area shows these routes’ full extent. This scale doesn't name every stop, just the general neighborhood destination (including all tube/rail stations), each of which will have its own spider map when you get there.
It’s true that these maps don’t help you if you need to make a bus transfer, but London’s bus service is popular enough that they can provide many frequent point-to-point routes. (There are no free or discounted bus transfers.) WMATA’s less frequent and less direct service, together with the city’s stronger orientation to landmarks and the street grid, would require a modified approach. A few Metro bus stops have reasonable “where can you get from here” maps, but most don’t. Just about anything would be an improvement for these stops, where you little hope of easily discerning where a bus can take you.
by peter dunn on Aug 18, 2011 2:01 pm • link • report
by dcbridge on Aug 18, 2011 2:12 pm • link • report
by spookiness on Aug 18, 2011 2:18 pm • link • report
One of the coolest things about the spider map above is the way in which the bus stops are lettered on the inset diagram of the area immediately surrounding Victoria Station in the spider map above.
Although this makes it easier to determine which bus stops where when accompanied by the table on the map, it also makes it much easier to actually locate the stop, because each stop is marked has that same letter emblazoned on top of the signpost (and the signpost also usually has a smaller area map that shows you where the other nearby stops are, and which buses stop at them).
I ride the bus all the time, and I love it. It's also definitely true that there are a number of fairly small steps that WMATA could take to make the bus network so much better. (Side-rant: The new Metro social media gurus almost never respond to bus-related inquiries). Doing this requires rationalization of existing routes, re-spacing stops along busy routes, creating express routes where necessary, signal priority in problem areas, and consolidating nearby routes onto the same corridor to maximize service frequency.
Oh, and the individual bus line maps are really, really bad. They contain very few reference points, and no compass rose (which is important, because they almost never face north). Here's the map for the mythical A11 if you don't believe me.
Allow me to nominate the X3 as a particularly silly route. It's nowhere nearly as bad as the A11, but it only runs 7 times a day westbound, and 5 times eastbound; all along corridors that enjoy excellent bus service with very easy connections to the X2 and 90/92/93 routes, which offers the exact same trip. (And, why is the X3 bundled in with the X1? They're only concurrent for a tiny fraction of their length, which they coincidentally share with the X2 and X9.)
by andrew on Aug 18, 2011 2:44 pm • link • report
by busgirl1 on Aug 18, 2011 2:56 pm • link • report
by Matt on Aug 18, 2011 3:19 pm • link • report
The map lists stations by single letters, but the reference table in the top left corner of the map forces you to search for the single letter station from the ordered table of actual line numbers. I think it should be reversed: the table should be ordered according to the stations and then should list the lines that service those stations.
by Scoot on Aug 18, 2011 3:37 pm • link • report
by Phil LaCombe on Aug 18, 2011 3:50 pm • link • report
It’s a little tough to read, but this gives you the idea: http://www.designworkplan.com/wp-content/detail-maps-london-busstop-TfL.jpg
(Note the signs that also give you estimated journey times, very helpful!)
by peter dunn on Aug 18, 2011 4:19 pm • link • report
The spider map concept is great, too - but I think the bigger functionality comes from simply being able to narrow the lines down.
by Jessica B on Aug 18, 2011 5:35 pm • link • report
The spider map for e.g. Charing Cross does have the intermediate zones. It's very well done, but requires fairly high resolution to display:
http://www.tfl.gov.uk/tfl/gettingaround/maps/buses/pdf/charingcrossquad-2048.pdf
by Matt C on Aug 18, 2011 7:45 pm • link • report
by dcseain on Aug 18, 2011 10:16 pm • link • report
by Matt Eldridge on Aug 19, 2011 12:45 am • link • report
Different people perceive the world in different ways. In teaching, we've belatedly come to understand the importance of supporting many learning styles. Wayfinding should be the same way: there is no one best answer.
by David R. on Aug 19, 2011 9:31 am • link • report
You're right about the Victoria map, I'm not sure why TfL does that one differently. It looks like a few other main stations also omit the intermediate zone (Liverpool St, Paddington). It may be for simplicity's sake at busy transport hubs, or it may just have been made at a different time.
The resolution is good in poster size at the bus stop, usable on a computer screen with some zooming, and pretty challenging on a mobile device.
by peter dunn on Aug 19, 2011 9:58 am • link • report
The most useful are the maps that show a few-block range around that particular bus stop. This is essential when changing buses. What looks on a citywide bus map like a red line crossing a blue line can really involve a more complicated transfer you can only see on a zoomed-in map.
Enjoy the links.
Whole Maps
http://bit.ly/pqRXzo
http://bit.ly/oPnAZq
Zoom-ins to see one intersection
http://bit.ly/r9o5hW
http://bit.ly/qxMnbH
Bonus: other maps Paris does better than we do
http://bit.ly/oZUySJ
http://bit.ly/pC3w78
by Josh Gibson on Aug 19, 2011 4:15 pm • link • report
1) In addition to the problem of knowing which buses go where is knowing at what times they run. The practice followed by most U.S. transit agencies of posting a single schedule listing only the major stops at every stop is extraordinarily unhelpful.
In most European cities, each stop has information like the following:
Line 32
M St and Wisconsin (northbound)
Time to: > Wisconsin and Calvert (5 mins) > Wisconsin and Macomb (10 mins) > Wisconsin and Albemarle *Metro* (18 mins)
Hour | Minutes
07 | 05 30 55
08 | 05 25 45 58
09 | 05 15 25 35 45 55
10 | 05 15 25 35 45 55
11 | 05 15 25 35 45 55
12 | 05 15 25 35 45 55
13 | 05 15 25 35 45 55
...
etc.
This easily conveys the time it takes to get to major stops along the route. And, given the current time, makes it easy to tell when the next bus is coming. I'm shocked that this has not caught on anywhere in the U.S. yet.
2) I would LOVE a mobile app that distilled the spider map concept to tell me what bus lines are around me.
For example, it would show only the bus lines that have a stop within half a mile from where my GPS say I am. For each one, it would also list the times at which the next TWO buses are arriving, and whether the bus goes to a Metro station.
by Andrew on Aug 21, 2011 9:25 am • link • report
In the case of WMATA maps they needs to go to the old style of maps (not having all routes red)
Another thing that should be done is to stop funneling all routes on the same streets downtown. Most of the bus routes in downtown DC travel along Penn Ave, H Street/I Street, K Street, 7th Street, 11th Street, 16th Street and Conn Ave.
Some routes should be changed to not all travel on the same route such as the 30's, 50's, 60's, 80's, 90's, A's, C's, D's, E's F's, G's, H's L's, Q's, R's, S's, V's, W's, X's, Y's all that start with the same number or letter travel the same route except for minor diversions.
We should redesign the system as a whole
Bus Schedules : the places listed on the schedule should be marked on the included map how they are in Ride On schedules.
Bus Stops: better info on what buses stop at each stop, where they go and when they run. Some stops list routes that have been discontinued for over 10 years.
Nextbus
1 the voice recognition is to sensitive it mess up whenever there is a policecar, fire truck, carn horn, thunder or someone speaking in the background.
2 Place monitors or e-ink screens at the bus stops to provide information.
3 Doesn't work during delays, detours
4 Should be able to provide alternative ways of getting to somewhere when a there is a detour
Map : provide different types every type suggested should be offered. Some of the maps are wrong; I have seen some that are geographically inaccurate and others that have wrong street names.
WMATA should build or have someone/entity build an application for computers/web/phones that has all the map and schedule info in it and is update that shows all stops on each route.
@ Andrew
how is that schedule layout any better it will cause confusion due to WMATA using the current system for 20 plus years.
@ andrew not sure if you are the same as Andrew that is capitalized
the X3 goes to Mclean Gardens which the 90,92 nor 93 the bus that does is the 96. Speaking of the X's and 90's there use to be alot more of them there used to be x1, x2, x3, x4, x5, x6, x8, x9 and 90, 91, 92, 93, 94, 95, 96, 97, 98, 99.
Most of the 90 lines changed when the Green Line opened and the X buses are always being discontinued and brought back the X1 and X3 were discontinued three or four times since the 1980's
by kk on Aug 22, 2011 12:50 pm • link • report
@kk, the difference between the schedule format I suggested and the ones WMATA currently uses is that they are customized to each stop. In other words, each stop would have an exact schedule of when the buses are expected to be at that stop. It also makes the times much easier to read.
In contrast, the current system requires users to figure out approximately how far they are between the two stops nearest to them as shown on the schedule, which requires a fairly advanced knowledge of the geography of the area. In addition, the text is so small and there is so much data that it is unnecessarily difficult to find a locate your current time.
The format I propose is not the only way to represent timetables at stations, of course. The main point here is to customize the timetables to the stop. Even if the times appeared exactly as they are now, printed in a single column, it would be vastly better as long as that column represented the bus stop you are standing in and not one down the street.
Yes, this would require WMATA to maintain and update a distinct set of schedules at every bus stop. But they already have that data in their computer system anyway, how would it really be to print out a unique set of schedules for each stop, then drive the length of each route once to put them up?
by Andrew on Aug 22, 2011 1:01 pm • link • report
How does it effect the info such as a person wanting to know how it takes to get from point A to B or to A to C.
How do you explain something of a change on this level to the masses.
What about NextBus why not just go the digital route and add a 5 to 6 inch screen to all stops showing when the next bus comes.
Time to: > Wisconsin and Calvert (5 mins) > Wisconsin and Macomb (10 mins) > Wisconsin and Albemarle *Metro* (18 mins)
Hour | Minutes
07 | 05 30 55
08 | 05 25 45 58
Why have the need for people to do addition when they could just look at the current one and what time it would get there it waste more time especally when the time does not fall on a 0 or 5.
Most people are stupid when it comes to Math so including addition and subtraction which may occur for places that change route certain time of the day should be avoided at all cost.
The goal should be lot of info with short amount of time in a clear manner.
Why not combine the schedule you suggested with the current one
Large bold Blue or Red 7 : followed by the minutes smaller
05
30
55
and have times for other places to the right in small print its a win win.
by kk on Aug 22, 2011 2:04 pm • link • report
I have one thought I haven't seen from anyone else: Metro needs these maps up on the street at the Metro entrances. It's crazy that WMATA thinks these maps are useful underground instead of on the street.
by david on Dec 8, 2011 2:26 pm • link • report
Add a Comment