The importance of frequent transit service

I was flying back into the SFO airport yesterday, and since I live in the South Bay, I needed to find some form of transportation to get me home from the airport. Fortunately, there is a BART station right at the airport, and I live quite close to a Caltrain station. I even had my Translink card with me. Unfortunately, I missed a BART train by 30 seconds. Since BART only runs every 20 minutes, that means that I would miss the connection to Caltrain, which runs once an hour, which would mean that I would get home an hour later than if I had run slightly faster to catch that train. Instead, I took a SuperShuttle, which, with waiting, actually got me home almost exactly when I would have gotten home had I not missed that train, but at considerably greater cos ($40 instead of $8.25). In this case, the reason I did not take transit was not transit's speed, and that is given that a local Caltrain is pretty slow and waiting for the connection takes time too. Rather, it is end to end travel time, which is a function of both travel speed and waiting time, and the latter is largely determined by frequency. And it's a virtuous cycle: increased frequency can increase ridership, and I suspect that the Moscow Metro would not be as popular if headways at 10:30 pm were 20 minutes instead of 4.

Translink on Caltrain: so many frustrations

Here's an email that I recently sent to Translink (soon to be Clipper) customer service to express my frustrations with the implementation of this fare payment system on Caltrain:

I've recently seen flyers stating that Translink will soon be the only form of payment accepted on Caltrain for monthly passes and zone upgrades. I feel that this is premature, and that there are a number of features Translink needs to have before it can match what is offered by the current Caltrain fare system. The main issues are zone upgrades, monthly pass activation, inter-agency transfers, and availability of vending machines.

First off, the availability of Translink vendors in the Caltrain service area is spotty at best. Right now, I can buy a monthly pass or 8-ride at any Caltrain station, whereas I can only do Translink transactions at the San Francisco or San Jose ticket offices during business hours, or at some Walgreens that are not necessarily near the stations, and may not necessarily even be willing to deal with Translink transactions (I'll file a separate complaint about the concrete cases of that). It would be nice to have Translink Add Value Machines installed at Caltrain stations, especially San Jose, but possibly also other popular stations such as Mountain View, Palo Alto, Redwood City, and Hillsdale.

The next issue is monthly pass activation. According to the brochure, when I buy a monthly pass for a given set of zones, such as 3-4, it is not activated until I tag on and off in the respective start and end zones, and that if I tag on and off in different zones, it will instead give me a pass for those zones (and I will be charged accordingly). This works great for someone who always travels between the exact same stations every time and never ventures anywhere else, but is a major inconvenience if I happen to be travelling to a different zone on the first day of the month. It means I either have to jump off the train at an intermediate stop to tag my pass, or else save the tagging of the pass for a different day and pay full fare, which seems incredibly unfair given that I've already paid for the pass.

On a related note, the handling of zone upgrades seems subpar. If I have a monthly pass, and wish to travel outside its zones, I have to buy a paper zone upgrade. If I do the intuitive thing and tag on and off, I get charged full fare, which is effectively charging me for something I already (mostly) paid for. Furthermore, in a comparable situation on AC Transit, it works the opposite way. If I have a local pass and wish to upgrade to a transbay ticket, I must use Translink e-cash to pay, not actual cash. This seems frustratingly inconsistent for a fare system that was supposed to provide some consistency among the Bay Area transit agencies. Also, with Translink, zone upgrades are no longer usable with 8-ride tickets.

Finally, there is the issue of broken interagency transfers. At San Jose, Caltrain connects to express buses operated by Monterey Salinas Transit and Santa Cruz Metro, neither of which are members of the Translink consortium. Both of these agencies provide discounts to holders of Caltrain monthly passes (in fact, MST provides free rides throughout its system). With the switch to exclusive use Translink as the medium for monthly passes, it will be impossible to provide these transfers unless MST and Highway 17 Express operators have the ability to read Translink cards.

I believe these issues need to be resolved before Translink can be made the exclusive form of fare media for monthly passes and 8-rides on Caltrain. I think Translink is a wonderful idea, and can be wonderfully convenient if implemented properly, but very frustrating if the implementation is half-baked. I hope these issues are addresses, because I'd very much like to see the Translink system succeed.

Privatization

From a report made by a rail construction company in the UK:

The move to privatisation resulted in a massive loss of skill and expertise at all levels in the rail industry. In many cases, the people who were lost were the people who set the standards that form the basis of what is in place today. When these people moved on they took with them the
corporate memory which formed the decision making criteria of what was done and why. The corporate memory issue is further compounded by the disaggregation brought about by privatisation with no one body holding all the information.

So much for the idea that privatization always makes things more efficient. And this coming from a company that almost certainly benefited significantly from it. For more on this, see The Navigators, a movie about the impact of privatization on railway workers.

A pleasant surprise from the MBTA

By far the most ill-conceived transportation project in Boston was Phase 3 of the Silver Line, a proposal to link Phase 1 (a silver-painted bus down Washington St) to Phase 2 (the bus tunnel from South Station to the waterfront), for no particular reason that was obvious to anyone other than that these two completely unrelated projects were called "Silver Line". Building a bus tunnel would have cost over a billion dollars required tearing up part of the Boston Common and demolishing an out of service Green Line tunnel that coincidentally goes to the same general place, and has the advantages of being a train tunnel and already being there. And nobody ever made a very convincing case that there would be any significant ridership from Dudley to the South Boston Waterfront (as opposed to, say, Park Street). Well, there's some good news.

The MBTA and the state government have officially endorsed a different extension of the Silver Line, a much cheaper and more useful proposal to extend the Washington St. Silver Line to a surface terminal at South Station and convert the #28 bus into another Silver Line-branded bus rapid transit line. While still not ideal (doing both routes as Green Line branches would have been better), this project is at least both useful and within the means of the MBTA to actually build, and is a huge step in the right direction for the MBTA planning process. Now all they have to do is admit that light rail is in fact sometimes better than a bus :)

Ever wanted to build a railroad?

Have you ever wanted to build your own railroad, but had no idea how to do it? Well, you're in luck, because thanks to the internet, you can now get the design standards from various railroads, including Caltrain and theDenver RTD (which has standards for both Light Rail and Commuter Rail). It's pretty neat to see how it's actually done, although a little bit disturbing that, for example, Caltrain's choice of transition spirals is based on what AutoCAD supports, rather than on the copious research both in the US and elsewhere about what actually works best for reducing track forces and improving ride quality.

Good work, MBTA

A few days ago, I took a ride on the MBTA Blue Line, and I must say, it's changed a lot since I last rode it, and mostly for the better. The collapsing platform at Wood Island is gone, replaced by a shiny new rebuilt one. And, at least on the day when I was there, all the trains were the new ones, with six cars rather than four. The new trains are, on the whole, very good. The ride is fast, comfortable, and relatively quiet, and the interiors are roomy. The only complaints I have are that the seats are hard, and that the drivers' cabs seem rather large, which adds up to a lot of wasted space in a six-car train with very short cars. But that aside, these trains are definitely an improvement on what came before, and are some of the best new rapid transit trains that I've seen.

Casino Express

Looks like the ACES train is finally about to start running. This is a privately funded train, organized by a consortium of Atlantic City casinos, and run by New Jersey Transit. Amtrak is in on this too: they're doing the ticketing. An interesting example of public-private partnership, the opposite of the more common one of public agencies contracting for service from a private company. And this is exactly the sort of service that public bus agencies are currently prohibited from providing.

CBTC just doesn't work

There has been a fad in the transit and railway world to abandon the conventional signalling technologies that have worked so well for the past hundred years, in favor of "Communication Based Train Control". This is a centralized system wherein, basically, every train constantly tells the central control computer where it is, and the computer tells the train how fast it should be going. Generally, these systems are highly proprietary, and often parts become obsolete soon after they're bought (San Francisco, for example, has a system based on OS/2). And the systems are generally very bad at handling vehicles that either don't communicate, or don't meet the criteria of the often-finicky systems.
As if that wasn't trouble enough, some bright fellows decided that rather than use the proven but expensive inductive loop technology, they'd use radios to communicate with the trains. Inside tunnels. Which tends not to work so well, and almost all of the systems where they've tried that have failed, including the Jubilee Line in London, the BART, and now Philadelphia, where the "more advanced" system has in fact reduced capacity and caused delays. Will they ever learn?

Everyone Loves the Electric (part n)

Amtrak has just posted their ridership and revenue statistics for fiscal 2008, and they're quite detailed and interesting. One thing stands out though: 55% of Amtrak's revenue comes from the Northeast Corridor, which carries 38% of the total ridership. If you do the math, you find that the average Acela customer pays $138, while the average Surfliner customer pays $18. For all that talk of how Amtrak needs to get rid of the NEC, it's still by far their most successful route, and the only one that has attracted large numbers of high-paying customers.

Why one should read EULAs

From the skype EULA:

Skype, in its sole discretion, may modify or discontinue or suspend your ability to use any version of the Skype Software, and/or disable any Skype Software you may already have accessed or installed without any notice to you

What this means is that at any point in time, skype reserves the right to hit the self-destruct switch on the software on your computer, and there's not a damn thing you can do about it, because you allowed them to by using it. Furthermore, they reserve the right to change the rules at any time by posting said changes 30 days in advance on their website, so it's your problem to check their website regularly. And best of all, this whole transaction is subject to Luxembourg law, and if you want to sue Skype, presumably you have to do it in Luxembourg.

The insanity that is VTA

The closer I look at the VTA (the local transit authority), the less I like it. The VTA seems singularly focused on wasting money in the most gratuitous way possible, getting the least bang for the most buck.

The epitome of this wastefulness is the BART extension project, which just never seems to go away. It's one of those things that seems like a good idea... until you think about it and realize that it's $6 billion and a 20 year delay for a subway from Fremont to San Jose,
through the middle of mostly nowhere. It's the same cost as LA's Subway to the Sea, and for two orders of magnitude less money, the VTA could have been running a commuter rail to Fremont or even Oakland, and in fact the VTA almost did before the BART project came along. But BART is still in the future, and only a relatively small amount of planning money has been spent on it.

The VTA has already wasted hundreds of millions of dollars on misguided projects, such as the light rail system. They sold their 15 year old fleet of 50 high-platform cars and bought a brand new fleet of 100 low-floor cars, despite needing at most 61 of them for peak service. And to allow level boarding with these new cars, they have had to raise the platforms at every single station in the system, at great expense, not to mention inconvenience to passengers when the
stations are closed.

And the design of the light rail system as a whole just seems wasteful: it doesn't really go where people want to go, and it takes a long time getting there. Just look at the absurd shape of the line from Downtown to Alum Rock via Milpitas. Or the long, scenic, and slow windings of the Mountain View line through office-park-land. Or the sidewalk running through Downtown at 10 mph, which makes trips through San Jose infeasibly slow. Or how the single busiest transit corridor in the county, along Santa Clara St and Alum Rock Ave, still doesn't
have light rail service.

Trains and scalability

What with the high gas prices, it's not surprising that even in LA, public transit ridership has been setting records this year. What's particularly interesting about this latest set of numbers, though, is that the Gold Line's ridership has finally exceeded that of the Orange Line busway, and the Orange Line's growth is starting to stagnate. There's a very simple explanation: the Orange Line has basically reached capacity, since it's going to be hard to maintain reliable service with headways closer than four minutes. On the Gold Line, they've gone from 12 minute peak headways down to 7.5. Plus, thanks to the ability to couple trains, as well as the higher speed that comes with proper grade crossing protection, the Gold Line requires 11 drivers to operate peak service, while the Orange Line requires about 24, for a roughly comparable number of passenger miles.
And the Gold Line has quite a bright future ahead of it, as higher ridership leads to a more frequent and thus more attractive service. The Eastside extension can only help matters, by linking more destinations with the rail network. With some upgrades to the power system, they will be able to start running three car trains and boost capacity by another 50% with no extra drivers, and I suspect that headways as close as 5 minutes are possible as well, more than doubling the line's capacity.
Meanwhile, the Orange Line is already running a bus every four minutes at peak, and it's not likely that service can increase much beyond that without creating jams at the many grade crossing on the line. So the buses will get increasingly crowded until an equilibrium is reached between people needing to get somewhere, and people not wanting to put up with the crowding. Or until the line is rebuilt as light rail.

Yet another Bush Administration WTF

The Federal Transit Administration recently made a ruling that expands the definition of charter bus service to keep publicly funded transit agencies from competing with charter bus companies when they provide premium fare service, like to football games. However, this isn't really a charter service organized and paid for by one person, rather, it's a special transit service, with the bus company doing the organizing and each rider paying an individual fare into the farebox. And according to California regulations, this makes it legally a transit service, not a charter service, and a license to operate charter service isn't enough to provide it, and it can only be operated by a transit provider. The end result might well be that nobody is allowed to operate this service at all. WTF.

Mixed-use development

These days with high gas prices and awful traffic, we are starting to hear more about mixed-use development, generally with terms like "urban villages" and such. The point seems to be that people can live next to where they work and have short commutes. But that is, for the most part, just wishful thinking, and entirely misses the point of being in a city, that being the access you have to a wide variety of jobs and leisure activities. A city is more than just a bunch of villages concatenated together.

No, the real point of mixed use development is just plain old efficiency. Things like raods, transit, and sewers are expensive to build, and they have to be built to meet the peak demand. The neat thing about mixed-use is that the peak demands of the uses generally come at different times, so the overall peak is only as big as that of the biggest use. If you've built roads and parking lots to support an office park, you can probably build a shopping mall next to it with only a little extra road capacity and parking, since the office park is empty on evenings and weekends, which is exactly when the shopping mall is full.

Carpets in subway car

I was reading some documents published by WMATA (which runs the Washington Metro) about their subway train fleet, and I learned an interesting fact: 15% of maintenance downtime is due to carpet replacements. I suspect this is slightly less bad for BART, but it's still pretty significant. No wonder both systems are looking for ways to get rid of carpets and save themselves money and maintenance headaches.

Another interesting fact is that 20% of in-service delays related to equipment failure are due to the ATC system. So maybe there's some advantage after all to having a nice simple mechanical train-stop system like NYC does: it's very, very reliable, and most light rail systems can get by just fine with no train-stop at all, just plain old block signals, again with much higher reliability since there's no carborne equipment to fail.

Automobile

I read a post on a forum on the topic of cars, and I think it is a wonderful demonstration of the sheer cognitive dissonance that is America's "love affair with the automobile".

Nothing will ever beat the private car for convenience. Its right there, whenever you want it. Its fast, it can be used by almost anyone, regardless of physical health. No wait times to use it, no sharing it with the smelly unbathed guy, the psycho homeless person, or the screaming infant. No stops along the way. And it can be used for trips of any length, to any location, without being forced to walk a mile from a bus stop to the destination. And depending on where you're driving, it can be quite pleasant- driving in the mountains with the top down is *fun*. I've never had a fun bus ride.

Apparently this person has never lived in a city, never had to walk a quarter mile across a parking lot, never had to sit in traffic for half an hour, never even stopped at a traffic light. He's never gotten drunk and been unable to drive, never been a teenager too young to drive or had a grandparent too old to drive safely.

Bofa?

When I came over to do my internship, I decided that it might be convenient to have an account at a local bank, so I opened myself a Bank of America account, since it was free. But then I got the pamphlet explaining the fees, and it's just... horrible. They charge for using another bank's ATM (and of course the other bank charges you too). Even to check your balance. And to make up for the fact that ATMs don't charge for using foreign cards, BofA charges even more for using a foreign ATM. And just today they sent me a letter explaining that they've confiscated my paycheck for a week, basically giving themselves an interest-free loan. It's like they think that I should be paying them to take my money and make a profit with it or something. I find that sort of attitude in a bank to be most... unhelpful.

Abonnement général

Switzerland has one of the best train systems in the world: It may not be the fastest, but it is one of the most comprehensive and reliable. You can get from anywhere to anywhere else in the country, trains almost always run at least hourly, and generally scheduled for easy connections. And while it's not cheap, if you live there, you can get all sorts of discounts, and they even have a yearly unlimited pass. That's right, if you pay 3100 Francs (worth about $2500 before the dollar became worthless), you can ride on any train, bus, tram, etc. anywhere in the country as much as you like.
Imagine if such a thing existed in Northern California: you would be able to go from Salinas to Reno, from Yosemite to San Francisco, all for just one easy payment of, say, $3000. That's hardly more than a year's worth of Caltrain and Muni passes, yet it offers one the freedom to go anywhere at all, without having to bother with tickets, or even think about how much it costs. You just hop on the train and go.

Automotive abomination

As I was walking down the street here in Sunnyvale, I saw a giant one-and-a-half-decker RV, towing behind it a trailer with two jet skis. And then it pulled into the gas station where gas is $4.20
a gallon. I can imagine the many-hundred-dollar fuel bill this guy had, and I can't imagine that this sort of fuel-guzzling wastefulness can keep going for much longer, with gas prices as they are.

Riding the VTA

Now that I'm in Silicon Valley, and temporarily bikeless, my main form of transportation is the VTA. While the buses and light rail are nice, and there's at least some service in most of the area, the system as a whole is just too slow, infrequent, and doesn't go where you want it to. For example, there is no bus that goes from downtown Mountain View to Google. Likewise, there's no bus that goes from where I am to downtown Sunnyvale: I have to take the light rail two stops and then transfer, waiting another 10 minutes in the process, and making a 3.5 mile trip take over half an hour in the best case. Another example: my friend wanted to go to a rollerskating rink in southern San Jose from Mountain View. This rink is conveniently not far from a light rail station, and Mountain View has light rail too, so one might naively think that this is one of those rare trips that is well served by transit. But no, it takes 1h40 from station to station, versus about 40 minutes by car. Taking Caltrain from Mountain View to Tamien and transferring there is generally much better, taking 53 minutes station to station, but that means you have to catch one of the few Caltrains that actually stop at both Mountain View and Tamien, and on the way back, Caltrain is no longer an option, as the last one leaves Tamien at 9:23, and the last train from San Jose is at 10:30 pm. It's just not practical: you can do it, but you'll feel like you're fighting the system the whole way. But I think there's some hope for the VTA, and maybe one day I'll explain what I think needs to be done.