CTA Tattler

Nice job-- they finally implemented what RTA designed tested and developed for them 5 years ago. Now how bout realtime signs/data on the rail system to integrate with the businfo?

As a note: these maps use Java which is entirely different than Javascript. As a comparison, Google Maps uses Javascript. It's a bit semantic I understand but it's important to distinguish them. With a Java based application the system is locked down and unavailable in any other format. However, the text based version while not technically "open" is still open enough. Take for instance Route 20. You can actually bookmark your bus stop and direction and by hitting refresh see up to the minute information.

In theory, someone could write a script to "scrape" the information on these pages and then create an open format (such as XML) which could then be used in another way.

It's great to see the CTA taking these step although I look forward to the time when the information is made available in an open format. Oh, and sorry for the geek talk.

Ah! I take it all back. The estimated Arrival times uses Javascript and updates every 60 seconds. Unfortunately, the data is still locked away. The update of the time and weather is a nice touch.

Caught this in the Tribune:

Google, CTA teaming up to aid trip planning:

CTA customers can now use Google to plan their rides on trains and buses under an agreement announced Tuesday.

Story here
http://www.chicagotribune.com/technology/chi-cta-google-webapr09,0,3820867.story

Google link at http://maps.google.com/chicago

Wouldn't scraped info be too out-of-date to bother using?

And, irony of ironies, now that the CTA finally has implemented what the RTA kicked off, the RTA has declined to participate in the finished product, so no directions on Pace or Metra.

The more I see what is going on, the more I am beginning to think that the Huberman CTA is going to end up as the model agency and Pace and Metra as the basket cases. How about merging all three into a single agency?

Nice. Too bad the CTA, in its (utter lack of) wisdom, decided not to provide tracking service for any of the bus alternatives to the severely-hampered SB Red Line. I would have liked to have seen at least one of those buses on here--like the #22 or #156--but lord knows I'm not holding my breath.

Ed and other developers interested in building services with Bus Tracker's data:

I've created a Google Group where we can discuss these efforts in a more thorough and geeky fashion. I've posted some of my observations, which hopefully will prove useful.

URL: http://groups.google.com/group/ctadevelopers

In plain English to all others reading this: By making Bus Tracker data available in an open format, it'd be possible to build a variety of third-party apps and services that would make Bus Tracker even more useful. Imagine things like widgets for Mac OS X or Windows and customized interfaces for the iPhone, for example. Like the new partnership with Google Maps, all of this can be done at zero cost to the CTA.

I'll second the inclusion of the #22 and propose the #36 broadway as well.

KevinB

I third it... all opposed? Motion carried! The CTA will include the #22 and #36 buses immediately!

Is that CTA immediately or real life immediately?

Or is it KevinB immediately?

It was fun to sit and watch the map. Look at the right time... or perhaps just about any time, and you can see bus bunching in action.

Kind of makes you wonder what made the first bus in any bunch late.

They should make the bubbles identifying each bus smaller so buses crossing each other don't cover each other up so much. Lots of blank space in the bubbles that's wasted.

Hey, I just watched the tracker confirm one of my suspicions -- bus bunching happens in no small part when drivers hang out and shoot the sh*t at a terminus and then take off a moment or two apart. On the 49 Western southbound route, bus 6553 left Berwyn & Western about 90 seconds before 6546. They're now in Lincoln Square a few feet apart.

Unfortunately the tracker's not completely compatible with Safari -- I can't display the schedule using the pop-up menu -- so I can't see when each one was actually supposed to leave. But they left Berwyn & Western about 10:20, again less than two minutes apart after both sitting there more than 10 minutes. I wish this had run numbers so I could turn snitch.

Well, checking the CTA's own site showed me the 49 Western schedule: One bus was scheduled to leave at 10:06 PM, the other at 10:19 PM. So one was (and still is) nearly 15 minutes late only because the two drivers decided to have a chat at the terminus and keep riders waiting on a cold, thunderstormy night.

FWIW, I think the text-only interface could be much improved and simplified. The first step is to choose the route, and then to choose the direction. Once I choose the direction, why not show me a list of all buses headed in that direction and their current location? That way if buses are bunched 6 miles away (as those two buses I mention above are), I can grab a cab. Rather than make me choose a specific stop, show me the lay of the entire route so I can make a quick decision. That would bring the text-only interface up to par with the route display.

You're making quite an assumption as to why the bus left the terminal late.

For example, labor laws regarding breaks apply to CTA drivers, too. They're just as protected as any of us. If that bus was late getting to the terminus, the driver is still legally entitled -- actually legally obligated if you ask some folks in the labor department -- to take his/her full break even if it puts him/her behind schedule.

Of course this doesn't help the passengers waiting out in the cold, and also sets up all kinds of nasty problems for at least the next two buses on the route, but (depending on why the bus was late getting to the terminus in the first place) may not be the fault of the driver.

Everyone wants to have the highest level of efficiency, and that certainly must have resulted in someone deciding to cut recovery times at the end of each trip. (I'd be shocked if average scheduled recovery time today isn't less than it was ten, or even five years ago. Certainly some penny-pinching manager earned some sort of bonus for cutting them -- perhaps multiple times -- over the years.)

Of course it's only just another assumption that in any given example that too short recovery time was the cause. But it's no less valid assumption than assuming that the driver spent extra time socializing. And, of course, even if a late arrival at the terminus was the cause of the late departure, that still doesn't answer the question of why the arrival was late in the first place. And this is where supervisors who have access to more data come in.

And drivers should need to make a note every time they fall more than a couple of minutes behind schedule. I'm not talking about a report as verbose as I tend to be, but they should be required to note things like trucks blocking, trains crossing, more than normal wheelchair ramp activations -- whatever caused them to fall behind.

Would drivers be truthful? For the most part, I think so. But when patterns show up, that's where the supervisors come in. They can investigate any patterns to see if the pattern justifies a schedule adjustment, or whether there's some driver issue to resolve.

My point is that thanks to the tracker maps we can see just how bad bunching can be. But we don't have enough information to start laying blame. What we can see is how the people who are in charge of fixing these issues now have a better tool to use in their job.

After all, the primary reason for the tracker isn't to give us another toy. The primary reason is to give supervisors better tools to do their job. Us getting a toy is just a secondary advantage.

I sympathize with the people from the North side who wanted this to help during triple tracking.

But I think it was a good idea to try it elsewhere first and get a better test than the 20 Washington was able to provide, in a spot where there are fewer internet/handheld users.

If they had implemented this on the 22, 36 and 134-148, and there were significant problems, that harmed thousands of people, or even if there were just bandwidth issues that it was difficult to predict based on the usage for the one existing route, we'd have a full-fledged mob riot.

It's too bad they couldn't roll this out as quickly as they had announced last year, but that's water under the bridge. I look forward to more details about the expansion they announced for May. Have they said what garage will be second?

With ya, Jake. The point of a rollout is to shake the system's bugs out, not to stress them out.

And Rusty, I don't understand your claim that "the bus was late getting to the terminus in the first place." I make it pretty clear it wasn't.

No, Bob. You didn't make it clear. You said they both sat there for more than 10 minutes, but if that first driver was suposed to take a 15 minute break before leaving, nothing you said makes it clear that they took more than their 15 minute break. He/she could have arrived late to the terminus, and left before the 15 minute break was over, and it would still fit the scenario you outlined.

What time were they suposed to arrive at the terminus? What time did they actually arrive? How long of a break were they required to take? How long of a break did they actually take? Saying "after both sitting there more than 10 minutes" doesn't address any of this, let alone make it clear that the bus didn't arrive at the terminus late in the first place.

I'm not saying that the bus was late getting to the terminus, but you don't know that wasn't the case, either.

A solution needs to be found, but jumping to unsupported conclusions about the root cause isn't going to help. You've got to know what's wrong before you can fix it.

Rusty's right, Bob--you didn't make it clear. You said nothing about when the buses arrived at the terminal, just that they left less than 2 minutes apart.
Could it be that one is an X49? (Long experience with the X80 tells me both 80s can be leaving Broadway like that.)

You're right in that if the drivers are shooting the breeze while on the clock & ignoring their schedules, that should be addressed, nay, stopped, immediately. No argument there.

For routes like Irving Park and Western, which are served by both express and locals at certain times, printed schedules may not be used any longer. Rather, the buses are deployed based on when they reach the terminal and when the previous bus left. Huberman said last fall that they were trying this strategy because attempting to adhere to printed schedules was creating bus bunching.

What exactly is meant by printed schedules? I still see a lot of drivers on the 80 & X80 with the little paper attached to their steering wheels, and that's what they wave at you to show we're early when you complain about them sitting.

Fair enough, Rusty and Dee. I was watching much earlier and much later and those buses never wandered more than a block apart over two NB and two SB runs, so I'm going to assume I'm right circumstantially, but I can't say I personally witnessed them. (And at that time of night, there were no X49s.)

The cynic in me says that one of the reasons they didn't do the 22 Clark is that they'd see how horribly bad the service and bus bunching is there and how off schedule they really are.

On one positive note, with the electronic tracking, I'm hopeful that the familiar picture of the supervisor sitting in the drivers seat of his gas guzzling SUV with a pen and clipboard will be a relic of the past soon...

KevinB

Yep. The supervisor will have a video display and keyboard in the future.

The tracker tells you where a bus is, but it still is going to take real supervisors in the field to determine why that bus is where it is.

It would be nice if there were more than two zoom settings on the Bus Location Map. Not being able to zoom out farther makes for an awful lot of scrolling around.

Also, it's puzzling to hear that the CTA still needs to "work out the kinks" on this program. All that time they were "working out the kinks" on the #20 route weren't enough?

And how, after the lengthy, lengthy time that the system has been in place on the #20 route, can the bus bunching problems be as severe as they are?

For example, at 10pm tonight -- on a night with no inclement weather and at a time when there is little traffic -- there were no westbound #20 buses west of Halsted. There were *three* east of Halsted.

While I've never really taken seriously the idea that the bus tracker will eliminate bus bunching (the main benefit is information to riders), it's sad to see that the CTA really isn't even trying. (Or if it's trying, that it's failing so miserably.)

Why on earth would the CTA not run one of those three express for a while to even things out? They've finally started doing this on the rail lines -- although sometimes ineptly, skipping major stops rather than minor ones, but at least they're trying.

I am anxiously awaiting for this to hit the 147. Catching that bus on the weekend is always a crapshoot.

Today I used Bus Tracker to time my departure from work. It allowed me to minimize my exposure to pouring rain and teenagers at the end of the school day. I'm eternally grateful for both. So far, so good. Previously I'd used it only for sport. From my office window I have a great view of a major street that hosts both a local and express bus. It was too much geeky fun logging on to Bus Tracker and then watching the bus go by. What's on the screen generally matched up with what's on the street.

Hey, I can't believe this only just occurred to me: During all the blockages and reroutes this summer (expected from street fairs and unexpected from everything else), how will the bus tracker handle the info?

I assume the main bus location map will accurately display the location of all buses, but are the route displays (the smaller popup windows that just show a linear diagram of the route) flexible enough to handle changes, especially on an ad hoc basis (such as the "police actions" that we occasionally hear about over CTA Alerts)? If you use the screen that displays upcoming arrivals at nearby stops, will it steer you to the nearest stop that the route's using for that duration in real time? And how about the text-only display for mobile devices, which is going to be pretty important to people who realize their nearest bus stop is out of commission for the time?

Last summer during a street fair on Clark, I called the CTA number to find out how the 22 was being rerouted so I could take it south. The assistant who answered had no idea that there could be any kind of reroute going on and told me that if I wanted to try calling the garage, maybe someone there would know. That isn't promising.

Good question.

The systems in use in different cities handle this a little differently. Essentially they could be using straight GPS info, or they could be using GPS info plus additional input from the odometer. And I wouldn't be too quick to assume that the map will accurately show buses when they move off their intended routes.

Also keep in mind that what we, the public, gets to see may differ from what is available internally. So while an employee (such as a supervisor) hits the internal servers directly, they may be able to get true real-time GPS information on any bus. But it's unlikely that those of us hitting a webserver could have that kind of access. (Giving us that kind of access could potentially give us the power to crash the server too easily.)

What I'm going to guess is that the map will be updated as if the bus were continuing on it's regular route until the next GPS check-in, at which time we might either see the bus "jump" to it's actual location off the regular route, or perhaps disappear until GPS puts it back on it's route.

Same for the arrival and departure times. We'll see time for stops even in an inaccessable part of the route, only to see sudden updates that will make it appear that an approaching or arriving bus just disappeared.

But I'm only speculating.

Great discussion and ideas, Bob S. and Rusty. I wondered the same today on the SB 11 when it ran the red at Fullerton & Sheffield and screamed to a stop on the southwest corner instead of making its usual left. If and when the North Park buses ever get Bus Tracker, how would the software interpret the re-route necessitated by today's work at Fullerton? I imagine all those bus icons being superimposed at the initial point of re-route until they are freed once they return to their regular route. On an amusing note, today as we returned from the Sox game, my 4-year-old nephew announced that it's better when the Red Line uses the elevated tracks because, "the subway is dark and boring." I told him they were fixing the tracks so we'd get out of the dark and boring more quickly. He told me the subway is stupid and should be filled in. The wonders of slow zone elimination are lost on the young.