Orphan Road

As promised here is an updated on my independent study. Since I meet with Prof. Dailey in the CSE department my studies have mostly focused on Mybus and how the information that is already there can be better communicated and used by riders.

As I documented earlier I first focused on how print information both at the stop and in leaflet form could look, which can bee seen here and here.

Since then I have been focusing on what the online information should consist of and how it should be communicated. I decided that the online component needs a better way of intuitively and quickly finding the mybus information, which resulted in the google map that I created. Currently mybus stop ids are in a tabular form split over 8 different webpages, which really isn't too easy to use.

My map includes all 1020 mybus timepoints. Each marker is linked to the corresponding mybus website, which a user can then bookmark. The next step is to improve the arrival time interface to a more initiative countdown or graphic format. I have a good idea of what I want to do but I'm trying to figure out the technical aspects right now.

Thanks to those people that have gotten back to me already. Some people have said that the most useful thing would be information at the bus stop. This is an example of what that could look like.

Over the last few days I have been working on creating this map for my independent study. What I tried to do is combine a simplified system map with the Mybus SMS service.

I realized through my studies that transit information consist of spatial and temporal information. Thus you have a route map to show where the bus goes and you have a table to show when the bus will be there. Subway systems essentially eliminate the need to know temporal information because they come so often. This allows them to simply focus on the spatial aspect of the transit system which is much easier to understand. That is why everyone loves subway maps. Using Mybus SMS I'm able to do the same thing.

This map only covers downtown, capitol hill and the u district but if I get a good reaction I will make a city wide map. Please let me know what you think.

Cascadia Prospectus reports that there's a new study underway to change the way real-time bus info is captured. The current system involves having the bus pas a series of mileposts, which doesn't work too well in crowded areas or when the bus has to divert its route due to construction or snow. The new method would involve GPS and/or Wi-Fi, which "would vastly improve the tracking in the urban area."

Another part of the study would involve counting passengers. Perhaps the algorithm somehow uses the number of boardings to determine how many stops the bus will have to make, and uses that to calculate the arrival time.

This makes it clear just how tricky real-time bus information actually is to implement. With a train, it's relatively straightforward: trains don't get stuck in traffic and make regular, predictable stops. But even the most sophisticated GPS bus system can only tell where the bus is right now and then make a guess about how long it will take to close the distance between the bus's current location and yours.

In most urban areas, though, that's probably enough if you're a regular commuter. If the system can tell me that the bus is still a half-mile away, I can make a reasonable guess about when it will get to my stop. Also, if I have the option of taking the local or waiting for the express, all I need to know is how far behind the express is relative to the local. Exact times don't really matter.

CP also points to this 2003 P-I article on the Mybus pilot program in North Seattle, which goes deeper into the local v. express dilemma.

Hot off the presses. Transit information system NextBus will be coming to Seattle.

According to my sources, the soon to be completed SLU streetcar will incorporate NextBus technology.

Unlike Metro's homegrown MyBus (and the equally homegrown BusMonster), NextBus is already being used by transit systems around the country, including BART, D.C.'s Metro, and more. They've got some really cool functionality, including an online form that lets you set up automated text messaging to alert you to the impending arrival of a certain bus or train. It's great stuff.

While I'm all about the "can do it" spirit that's gone into MyBus (and thanks, Google, for the funding), I have to ask Metro -- why build something from scratch when there's already a great application out there that does everything you could want it to do?

NextBus suggests that we write Metro and the city and county governments to encourage them to adopt their system. If you're interested, here's the Metro "comments" page: Metro comments.