NextBus a big success for Metro

Metro has delivered a quality benefit to riders with NextBus. If it’s not a home run, it’s at least a triple. From my own experience and anecdotal evidence from others, if NextBus says a bus is coming at a certain time, it’s almost always there right then or a minute or two later. That’s a very valuable service.

For example, I recently used it for a bus that runs only every 30 minutes middays and stops a block and a half from my house. Normally, I’d avoid this bus, or plan for a 10-20 minute wait at the bus stop. Instead, I monitored the prediction, then kept working until almost the exact moment I had to leave. Then I started running a bit late, so I knew to run to the bus. Sure enough, the bus pulled up just moments after I reached the stop.

NextBus’s one significant flaw is the buses that don’t appear, either because the operator didn’t activate the transponder or because it’s broken. In those cases, a bus rider might decide to walk across the street to buy a sandwich only to see a bus glide past from the windows of the store. Metro has been training drivers to activate the system, and anecdotally, these occurrences seem to have declined.

According to a Metro presentation for this Thursday’s Board meeting, the Web site got 152,881 uses in July and 143,052 in August. The phone system’s usage rose in August (from 69,242 calls to 87,197 calls) despite the traditionally slow month, probably because Metro rolled out more signs on more stops. Metro will continue to deal with vandalized signs and replace broken radios. They also plan some customized management tools for them to analyze bus performance using the data, and will “consolidate the bus stop data base,” though it’s not clear what that means.

Metro only received 136 complaints in July and August. That’s great, and it’s a fine system. However, they use a very misleading pie chart comparing the numbers of complaints with the total number of people using the system, which gives the impression that every person who didn’t complain must have liked their experience. That’s obviously not true, even if the vast majority have been happy, as they probably were.

I’d encourage the Board to ask two questions:

What percentage of buses don’t appear on NextBus? If an operator forgets to turn the transponder on, the bus doesn’t show up in the system. Likewise, if the transponder is broken, it doesn’t appear. According to the presentation, it appears Metro ran some QA on this. What did they find? What percentage of operators aren’t turning it on, and what percentage are broken? Do they have a target standard for this? How quickly are broken transponders being replaced? How quickly does Metro find out that they’re broken?

Can other application developers get the position data? Tom Lee, for example, wanted to try developing an application using the position information. However, Metro doesn’t make it accessible via an API the way CTA does. Further, there’s some ambiguity about whether Metro, NextBus, or NextBus’s licensee “NextBus Information Systems,” which developed the iPhone app, owns the rights to the location data. Can Metro clarify the legal status of this data? Are there any plans to create an API so that those interested in building other types of applications can use the same raw data that NextBus gets from Metro?