visualising city transit layers

There are many layers to the city.  There's the city seen through its people, the city seen through its food, the city seen through its landmarks, and the city seen through its transit, amongst many others.

Our understanding of transit has been pretty static, access points connected by lines.  Go to this dot, be transported along this line, change at this dot...  Here's a section of Manhattan from Google Maps with the Transit layer enabled.

Map data ©2011 Google

(To get additional layers in Google Maps, click beneath the Map icon in the upper right.  Options include Traffic, Transit & Bicycling - some of the options require you to click "more" in order to see them.)

Or here's the complete MTA New York Subway map.

But computer power and storage, combined with new (ideally real-time) data, and new ways of visualising can give a very different view of the transit layer.

One of my favourites is TripTropNYC, which gives a "heat map" showing subway times to a given location.

For example, for Elsevier headquarters at 360 Park Avenue South you can see in an instant that there is a band of short travel times up and down along the east side, and blocks of short transit times even across the East River near subway stations (there are probably short routes from across the Hudson too, but TripTrop only shows subways and the subway doesn't go to New Jersey).

Map data ©2011 Google

(I don't remember how I originally found TripTrop, but there's a kottke article about it from May 2009.)

On June 5, 2011 The Economist wrote about another time-based transit visualisation

@TheEconomist The Economist

A new application shows you the places in your city that you can reach in a given amount of time using public transit http://econ.st/iN9Zpc

http://twitter.com/TheEconomist/status/77417110925606915

STEFAN WEHRMEYER, a 23-year-old German programmer, has developed a Google Maps application called Mapnificent (harhar). It's pretty cool: it shows you the places in your city that you can reach in a given amount of time using public transit.

Mapnificent: A time-based transit map - Economist Gulliver Business Travel blog - June 5, 2011

Here is Mapnificent's view of the same 360 Park Avenue South problem space, showing a 15 minute transit time window (the light coloured bubbles over the map)

Map data ©2011 Google

I have to wonder if it includes the Roosevelt Island Tramway though.  One of the challenges in urban transit is capturing ALL the different possible modes.

With Settings->Show color map checked, you get a TripTrop-like view.  With a live (if rather finicky) slider so you can see how the bubbles expand and contract as you increase and decrease transit time.

Map data ©2011 Google

Stefan Wehrmeyer's blog post about how it all works, A Mapnificent World, tells a key part of this story, which is open transit data... and licensing problems.

Data

Google specified the GTFS format in order to be able to use public transport information in Google Maps. The public transport providers publish a file with all necessary information inside and the magnificent GTFS Data Exchange collects these files and keeps them up-to-date in one place. So getting the data for the cities is easy. The hard part was that I have to read and adhere to the license agreement of every single provider. Some providers like New York’s MTA do it right by not having any complicated license text at all, some like San Francisco require me to put a text of 98(!) words on the site. I don’t know what they hope to accomplish, but I can work with that. However, I have a problem at the moment with the Terms of Use of Los Angeles Metro and San Diego Metro where they say that I’m only authorized to reproduce the data  in the form provided by them. What? Shall I show users the raw GTFS files? I guess not, but Mapnificent aggregates and converts data and I’m unsure if this is covered by the license. I have tried to contact Metro LA on two different channels, but did not get a response (their sites also make it extremely difficult to contact them by email). If you can change anything about that, please feel free to do it.

Ottawa provides GTFS. The local transit agency, OC Transpo, doesn't say much about it

Every service change (4 times per year), Scheduling and bus stop data is provided in zipped format to certain developers, including Google.

from http://www.octranspo1.com/about-octranspo/reports

Other than saying "certain developers" (while the file is open for anyone to download), it's not clear what license agreement applies, presumably the site-wide Terms of Use.  But Ottawa also has an open data site

http://ottawa.ca/opendata

Which includes more information and a pointer to the same file

http://ottawa.ca/online_services/opendata/info/transit_schedule_en.html

and with more generous Ottawa Open Data Terms of Use.

I mention this mainly for two reasons: 1) licensing is a big inconsident mess 2) open data sites are helping to fix this.

Here's a map.  NOTE: To switch to Mapnificent Ottawa, you actually have to click on the Mapnificent Ottawa icon, otherwise it will say "your point is out of the covered area".

Here's the 15 minute transit window to get to World of Maps at 1235 Wellington (Wellington and Holland).

Map data ©2011 Google

But again we see the river is a barrier - across the river is Gatineau in Quebec - which has a completely different bus service, the STO.  Some of the STO routes cross into Ottawa.  But while STO provides its routes in a PDF they are not available as GTFS as far as I can tell.

And the actual situation of urban mobility is much, much more complex than this.

Typical urban options include:

• walking
• cycling
• surface transit
• underground transit
• taxis
• car share services (Vrtucar in Ottawa)
• bike share services (Capital Bixi in Ottawa and Gatineau)
• car pooling including "just in time" matching of passengers to cars going near their destination
• traditional car rental and bike rental services
• helicopters, funiculars, aerial tramways, The Monorail and many more
• (sadly no dirigibles that I know of, although there really should be dirigible transit)

And there are all kinds of levels of complexity, where cycling encompasses everything from high-speed commuter cyclists who ride in traffic to more leisurely local cycling, and transit is a mix of commuter services with surface and underground stations (e.g. the RER trains in Paris and the coming Ottawa Light Rail) and local services (local bus routes and surface trams).

Onto this add the realtime layer: Where is the bus? Is it on time?  Is it faster to walk?  Is the path chosen for walking pleasant?  Safe?  Are there bikes at the nearest bikeshare station?  Are there empty slots at the destination bikeshare station?  What's the probability there will still be an empty slot by the time I get there? Is it likely to rain?  Does it pass through a location where many pedestrians or cyclists are killed by car & driver?

Here's a visualisation of Capital Bixi usage (which is only 100 bikes at 10 stations) - you can see stations fill and empty.

http://bikes.oobrien.com/?city=ottawa

Click Animation and Start Animation to see the bikes at stations change over time.

(You can use the pulldown menu to see other cities, e.g. Washington DC.)

Just to round out the story, remember that people may not have good time-based mental maps of their cities.  When London posted simple walking maps for Tube stations (with concentric circles showing the destinations you can reach with 15 minutes & 30 minutes of walking) some people discovered it was faster to walk to their destination than to take the Tube (here's King's Cross Station).

And remember that this requires not only transit data (ideally realtime) but layers upon layers of map data (also ideally realtime).  What are the walking routes?  Which are high-speed pathways for cyclists?  Which are recreational paths?  Is there road maintenance?  Has the sidewalk been plowed?  Is the road closed to cars to allow Sunday cycling?

Cities have so many different types of paths, like the Manhattan Waterfront Greenway, the Manhattan High Line Park, the Ottawa Capital Pathway - ideally all of these get combined together computationally into an integrated route planning picture.  That's a lot of data, usually provided only as PDFs generated from internal systems, from many different organisations, to try to somehow get into machine-readable frequently-updated integrated formats.

And layers of policy details: Can you put your bike on the bus? On the train?  On the subway? 

And layers of computation and user feedback.  Does the fast transit route put you on the opposite side of a busy highway from your destination, while a slower one drops you right at the door of the building?  Is this sidewalk covered with slush and ice?

And even a social layer - are there friends nearby or friends along the way to your destination?

We have fairly elaborate systems to track and react to car traffic flows, but basically zero to help pedestrians and cyclists flow around the city as conditions change minute to minute and day to day.  Ottawa's Traffic Operations Control Centre looks like something you would use to monitor the launch of the Space Shuttle.  But I suspect none of those screens display anything about pedestrians or cyclists moving around the city.

The ideal system would give lots of different options, incorporating all the different mobility scenarios - walk to a bike rental station, then ride to a transit node, take the bike on the train, the ride to a drop off station and walk the rest of the way - you need a lot of data working together to be able to construct these kinds of routes computationally.

Some elements to take away from this:

• having a data standard (GTFS) with downloadable files is already enabling innovation
• innovation is being stifled by incomprehensible licensing
• open data sites can help to centralise the discovery of files and provide a single clear license
• visualisations can help people understand time and movement through their city in ways that static transit maps can't
• multi-mode transportation requires surfacing the different durations and aspects of all the different ways to move around the city
• realtime data can add even more benefits
• we need better maps (OpenStreetMap can help here)
• we need better ways of feeding updates and opinions into the system (things like FixMyStreet.ca and OttawaBikingProblems.ca can help - if they feed into city 311 systems and are acted upon)
• we need mobility operations centres, not traffic operations centres
• we can benefit from this data in many different roles: as citizens, as tourists, as city planners, as reporters
• there may be great partnerships possible, for example cycling map open data from the city, transformed by cyclists/citizens/hackers to be the most relevant and usable for their real cycling experiences
• UPDATE 2011-06-07: while people can move easily across administrative boundaries, data does not; we need to make data integrate across boundaries ENDUPDATE

Two groups in Ottawa who look at these issues from a data perspective are @opendataottawa and @hackshackersott  To the extent that we get better at cracking this problem we will improve the urban experience for everyone - including local businesses and local technology entrepreneurs.