Transit
Map of the week: Personalized Metro travel time map
Ian Rasmussen created this travel time Metro map for a friend who was moving to Petworth:
Travel time maps are those that show places which take longer to reach as farther away. It follows the hand-made graphic design of Peter Dunn's Boston T travel time map, but lays out places in a way that resembles MV Jantzen's automated travel-time maps.
Ian writes:
Personalized maps could be the future
- It was made for a friend who had just bought a house in Petworth. Thus the center of the map's concentric time circles (each of which are one minute of travel time) is the Georgia Ave station.
- The travel times and headways were taken from the WMATA weekday peak schedules.
- The distance, in time, represented by a station where you transfer is half the headway time of the train you would be going to. This equates to the average wait time for your next train.
- The breaks in the lines are necessary so as not to violate the space-time continuum. What is shown on the map is the shortest travel time to any station (within 45 minutes of Georgia Avenue).
This map specifically applies to people whose home station is Georgia Ave-Petworth. Normally, we all use the same map. But why must this be so? In some places, it must be; there is only going to be one printed map on the wall of a Metro car, and a tour guide of DC will have the same maps for everyone who buys the book.
With technology, however, we could imagine maps being far more customized to the individual. My mental map of DC is different from another resident's. Mine has a lot of detail of the streets right around my house and not as much detail in a more distant neighborhood. Mine has certain Metro lines, bus lines, and bike routes I use frequently, while there are others I'd almost never use. Yet both get equal weight in the classic map.
Maps can't include everything, so they generally pick and choose based on modes or administrative divisions. The Circulator maps shows all Circulator routes and no Metrobus routes. There is a bike lane map and a separate bus map. None of this is on the Metro map. Sometimes that makes sense; if you're on a bike, you probably want to know the bike lanes. But you might also want to know about bus lines that have bike racks.
A map that shows the streets, places, bike routes and transit lines that you are likely to take, based on your own life, where you live and the places you go, could be massively useful. A hand-made map, like Ian Rasmussen's, Peter Dunn's T map, or Peter Dunn's H Street "spider map" can look amazing (especially that spider map!), but is a lot of work to make, while a computer-generated map still looks clunky. MV Jantzen's demo is great, but the lines and station names often overlap in ways that makes the map hard to read, because it's not easy to write software to avoid that.
There's no reason sophisticated-enough computer software couldn't do both, but it would take much more work which nobody has yet done. OpenPlans and others have talked about wanting to do projects in this area; if they can, one day people could be used to consulting entirely personalized and far more useful maps in addition to the ones we use today.
Comments
- Bikeshare is a gateway to private biking, not competition
- Judge denies injunction against closing schools
- Long-term closures: A solution to single-tracking?
- Metro policy for refunds after delays falls short, riders say
- PG planners propose bold new smart growth future
- Prince George's County struggles to get trails right
- M Street cycle track keeps improving, draws church anger







But that was the ideal map (spider one) I envision when I think of personalized ones. I really like the concept of this map, though I question the usefulness. If you're going from Petworth to Cleveland Park are you really going to take Metro? Probably not - you'll probably take an H bus or maybe bikeshare or something else? Like David said, if it's going to be a truly personalized map, it should / needs to include all options.
Finally, while this is nice (and I don't mean to get up on a soapbox) it perpetuates the stereotype that buses aren't a transportation option for 'newcomers' or tourists. We have a fantastic bus system and more people need to discover it. These types of maps - and esp the H Street spider one showed how - could go a long way towards helping get people to ride them and help relieve the Metro system.
by Shipsa01 on Aug 23, 2012 4:08 pm • link • report
by MLD on Aug 23, 2012 4:20 pm • link • report
I bike to work, and it only takes 20 minutes even though it's uphill each way!
This map a great concept!
by nettie on Aug 23, 2012 4:22 pm • link • report
It probably takes 45 minutes because you are counting when you leave your house (and look at the clock) to when you arrive at work (and look at another clock).
Pull it up on the trip planner - it's about 23 minutes not counting transfer time, so during rush hour probably 25-31 minutes. Again that's just depart to arrive, so it doesn't count getting from outside into the station, or back out again.
by MLD on Aug 23, 2012 4:33 pm • link • report
And great points about buses. It's so much harder to make a bus map than a Metro map, but the point about personalization is even more apt for them. While you might ride all 5 Metro lines from time to time, you almost surely will only use a small fraction of bus lines at all, and a map showing Metro plus those could be tremendously useful.
Peter Dunn's great spider map showed what could be, and it could do even more with personalization. But that requires a lot of work into developing software that can generate those, which is not easy at all.
by David Alpert on Aug 23, 2012 5:06 pm • link • report
But then again, that's not why I personally use maps. I use them for places I rarely go or never have been, or when I want up to date information of what is around me. Other things are so much better for learning how to get somewhere, the time it takes, and how much it costs.
Personalization to me is selective disinformation, and I would always question the source, just like it has been surveying me. "Eat here! You'll like it!". I can see utility in that if elegantly executed, but it does get exhausting always being the center of a universe. I think that's why I love of all types of maps but not so much the concept of personalized ones.
by andrewi31 on Aug 23, 2012 10:18 pm • link • report
MLD: If the "concentric circles" were actually circles you would need to adapt the transportation network to fit the time layer. You run the risk of deforming the network. In this way is much easier to establish an algorithm and make it personalized to everyone.
by Travel mapper on Aug 24, 2012 6:56 am • link • report
What I find most bothersome though is how limiting this winds up being regarding alternate itineraries. For the example shown of Petworth, it shows a path of Yellow/Green to Red to Orange as the means to get to Clarendon. Every time I make a similar trip to this, I will always ride to L'Enfant to connect to the Orange, even though it involves back tracking simply because it is a less complex itinerary and involves one fewer transfer and wait.
Last week. I was going from Pentagon City to Silver Spring and planned on taking the Yellow to the Red, but as I got there, the Blue line was pulling up and the PIDS display was showing 12 minutes for the next yellow. Hopping on the Blue instead of waiting for the more direct yellow got me to Metro Center in perfect time to connect to a Red Line train and get to Silver Spring in better time than waiting for the Yellow, even though my time traversing rails was longer. A similar map from Pentagon City would invariably show the Yellow as being THE gateway to getting the Red Line, with the Blue line likely stubbing to a truncation at around McPherson Square.
by Adam on Aug 24, 2012 8:08 am • link • report
MLD: If the "concentric circles" were actually circles you would need to adapt the transportation network to fit the time layer. You run the risk of deforming the network. In this way is much easier to establish an algorithm and make it personalized to everyone.
Right, but the base variable we care about here is time - this map is supposed to be based on a time layer and show us how much time it takes to get places. So the network SHOULD be deformed in order to show how the places relate on a time scale.
I get that it's easier to show it this way, but it also means I have to count rings every time I want to figure out how far away (in time) a station is. If the base layer of the map (time) were to scale, then you could just put lines at 5/10/15/20 minutes and be able to easily tell just by looking at the map the difference in time between the stations.
by MLD on Aug 24, 2012 8:11 am • link • report
I would personally add a time scale every 90 degrees here too. Network adjustment is indeed a better alternative but it would be much more complicated to implement. Who is up for the challenge?
by Travel mapper on Aug 24, 2012 9:08 am • link • report
by Don R on Aug 24, 2012 10:23 am • link • report
by PamD on Aug 24, 2012 12:16 pm • link • report
by Marshall on Aug 29, 2012 1:16 pm • link • report
Add a Comment