Slashdot Mirror
How Google Is Remapping Public Transportation
waderoush writes "Google wants to 'organize the world's information,' but there isn't a marketplace or a category of knowledge it can organize without remaking it in the process. A case in point: public transportation. Largely outside the media spotlight, Google has wrought a quiet revolution over the last five years in the way commuters get schedule information for local buses and trains, and the way public transit agencies communicate with their riders. GTFS and GTFS-realtime, which Google invented, have become the de facto world standards for sharing transit data, and have opened up space for a whole ecosystem of third-party transit app developers. This in-depth article looks at the history of GTFS and Google's efforts to give people information (largely via their smartphones) that can help them plan their commutes on public transportation — and, not incidentally, drive a lot less."
After missing three or four timed-transfer connections, I've given up on Google Maps for transit.
I'm sure it works sometimes, but since they've made it impossible to check their work (they don't give you access to the schedule data) it's a hell of a lot easier just to check the schedule myself.
That said it does work okay for short bus trips, but I've already got an app on my phone that tells me when the bus is arriving base on real-time data. No need to bring Google Maps into the picture.
There's no -1 for "I don't get it."
I'm a dispatcher with a small transit agency out in the Midwest on Google Transit, and I have to say its been great for us and our riders. New passengers are typically unfamiliar with locations around towns or unfamiliar with the local bus schedule, and giving them a trip planner that is already built into a familiar interface on Google sure makes life easier on them. The GTFS feed itself is also useful for external developers of programs that provide extra service to passengers, like Android or iPhone applications, or even members of the public that just want a well-documented view of exactly how the buses in a town operate. The fact that all of this is free is just icing on the cake.
A shout out to Bob Heitzman for his wonderful Excel-based tools (https://sites.google.com/site/rheitzman/) that enabled our system and others to get on to Google in the first place. Anyone out there who works for a small public transit system should check those out if you're wondering about supporting a GTFS feed. They aren't fancy, but they work well for outfits that don't have the manpower to run a full set of scheduling software.
Google Transit is not news to those of us who work in transportation. I work in Sustainable Transportation/Transportation Demand Management and my job is to get people to do (practically) anything but drive a car alone. Since I also work at a University, it's also my job to convince students not to bring cars to school (at least for the first few years) and it would be SO MUCH EASIER if I could convince Google to jump into multi-modal trip planning. Why?
Well, let's assume you're at my University and want to get somewhere 85 miles south without a car. You might be able to bus to the local train station, catch a southbound train, and then catch another bus to your final destination. However, the bus service here is contracting (sharply) due to budget constraints so a bus connection to a train will not always be an option.
I often suggest biking to the train, riding the train, and then biking to the final destination, but since Google Maps treats transit (bus/train) and biking separately, my suggestion can only go so far. It requires some rather involved planning for a novice to get from our campus to the train station by bike.
There are other options like OpenTripPlanner which, when coupled with a well-mapped OpenStreetMaps, can be an incredible way to plan multi-modal trips in addition to mapping out literally everything in an area from streets to bike lanes to sidewalks, stairs, and handicap accessible ramps... but it takes A LOT of work to perfect a local map and then to host an OpenTripPlanner server. It's relatively easy, but it's man-hour intense.
So, come on Google, pretty please.
It should be noted that Brian Ferris is from the United States. He got hired by Google to do transit work as a result of his thesis doing similar work for the Seattle area public transit (see OneBusAway if you are in the Seattle area and haven't heard of it). He's not some random European complaining about the United States.
Comment removed based on user account deletion
I've tried Google's public transport option a few times, but it's given bad plans. I don't know if they don't have good enough data, or if their algorithms aren't tuned to working in a city with so many frequent services as London, but the official route planner is much better -- and covers everything in London, which is enough for most people in London. Google's is very keen to switch from the Underground to buses -- it forgets that it takes much longer to get from the deep-underground platforms of the London Underground, crammed with people, out onto the street (which exit?), to the bus stop (which side of the road? which stop?) than to walk to a different platform for a different line. It's also optimistic with journey times during rush hour -- buses are often slowed by traffic/people, trains aren't much affected.
Also, when service frequency varies, it's most useful to know "take bus 23, buses are every 10 minutes" than "take bus 23 at 08:23". I've been approached by tourists concerned that there was no London Underground train at 20:42. Well, no, but there was one at 20:40, and another at 20:43.
Many European cities, and many elsewhere, have a single website with a routeplanner for that city. I expect many, many people still use these websites rather than Google.