It’s the beginning of a new year, so it’s a good time to look into the future. But here’s a caution for transit planners: don’t look too hard.

As a society, we like to believe in the power of technology to change things. Transit has not escaped that, and we regularly see new technologies touted as the solution to our transportation problems. A great, albeit somewhat dated, example above, from the excellent blog Paleo-Future: not only is it a monorail (a technology that’s still considered futuristic today, 106 years after the first urban monorail entered service) but it transforms into a ferryboat to cross rivers!

Here’s the truth: new technologies that truly transform transportation are rare. In urban rail transit, the last was in the 1880s and 1890s, when Frank Sprague perfected electric propulsion. If I spot you one non-transit technology — air conditioning — you could create an effective rail transit system using only pre-1900 technology. But don’t leap to the conclusion that transit is outdated: a freeway would be no less efficient using 1927 Ford Model As, and Southwest Airline could operate the same schedules using 1958 Boeing 707s.

But merely the fact that existing technologies work just fine has not stopped politicians and transit planners from trying to do better. The results have often been ugly. The most famous example is San Francisco’s BART. It was to be the subway of the future. It introduced a new computerized control system, a new computerized ticketing system, and even a new track gauge. The system had a cost overrun of 50%, and the new innovations never lived up to expectations. The ticket machines were temperamental, and the bugs in the control system took several years to work off. Even after service had begun, a train (luckily without passengers) ran off the end of the track and landed in a station parking lot. More remarkably, the computer technology didn’t actually improve on older technology. It turns out that it’s not the control system that determines BART’s capacity; it’s the track layout. Every San Francisco-bound BART train under the bay has to stop at a single platform in Embarcadero Station, and thus the time it takes for passengers to get off the train determines the spacing between trains. The Key System — which operated in the same corridor from 1939 to 1958, with the same number of tracks crossing the bay — had multiple San Francisco platforms, and it could actually operate more trains with a much more primitive signal system. BART is a useful system; it is hard to imagine San Francisco without it. But it could have been equally useful and been built in less time for less money had it used off-the-shelf technology,

BART’s experience is not unique. The Las Vegas Monorail was shut down when parts started falling off. The French GLT guided bus system, intended as a more economical alternative to light rail, has had repeated problems with vehicles swerving off the guideway, and it turned out to cost more than rail. Toronto’s Scarborough RT was initially buggy, forces a tranfer because of incompatible equipment, and is now proving enormously expensive to renovate because of its proprietary technology. The common theme is that these systems were chosen for the perception of modernity, not because of technical analysis. And the operators paid the costs of finding the flaws in previously untested systems. Buses and light rail may be boring, but we’ve built so many that we know what works and what doesn’t.

But while new technologies have not revolutionized transit as promised, new ideas have. Light rail was not a technological innovation; it was simply the idea of using streetcar technology but running the train in reserved right of way rather than in lanes shared with car traffic. The result is faster, more reliable, and more efficient services than streetcars or buses at a much lower cost than a subway. We can dismiss it as just a new way to use technology that already existed, but that’s true for Southwest’s frequent short-haul flights with fast turnarounds, or FedEx’s overnight service, or the unit trains that have made freight railroads so much more efficient. And the result of this idea has been better transit in dozens of cities.

There are surely innovations in transit’s future, perhaps in the use of computers to provide passengers with more up-to-date and more customized information, perhaps in new, fast faster construction techniques. But I suspect they will come with vehicles and guideways that seem quite familiar. There’s a real benefit to sticking with well-tested technology: more predictable budgets and schedules. In any project, public or private, that matters. And, by not focusing on technology, we can focus on what really matters: building transportation systems that go where people want to go and are easy to use.