An expansion of MARTA has been a long time coming. Too long for Atlanta to blow it.

A new law allows Atlanta voters to decide whether to raise the city's sales tax to pay for more rail and buses. Even for an agency that hasn't laid any new track in more than 15 years, however, there's such a thing as going too fast.

At the first of four public meetings about MARTA’s expansion plans, held May 25, Atlanta’s head of planning, Tim Keane, described the goals for expanding public transit.

“This is about creating more options for people,” Keane said, “and it’s about reducing congestion by getting more people out of their cars.”

Officials then listed all the wondrous projects that might be built. Not once, though, did they say how many people would be coaxed out of their cars.

Thinking I’d missed something, I inquired about the lack of figures. It wasn’t an oversight. The figures don’t yet exist, and might not before a November vote.

The explanation: Experts can’t project ridership, much less how many car passengers would be displaced, until there are more specifics about exactly where the new routes would run, where they would stop, whether vehicles would run in dedicated lanes, and how frequently vehicles would run.

Voters are being told this is merely “a call to implement,” as Keane put it, “a lot of work that’s already been done.” But what does it say about the work already done that it’s so vague no one can hazard a guess as to how many people would become transit riders?

As it happens, we have an example of what results from putting something somewhere just to do something: the Atlanta Streetcar.

You know the problems by now: Ridership is sharply below projections; costs are sharply above estimates. The state DOT has pledged to shut it down after June 14 if the city hasn't solved 60 problems it has known about, in some cases, since the misbegotten thing was launched at the end of 2014.

Against this backdrop of incompetence and even negligence, the city now proposes to add 30 miles of streetcar routes.

Well, maybe. You see, the total bill for the list of wondrous is about $6 billion. The total expected revenues? About $2.5 billion. So we would have to beg Washington to pay for part of it, which requires money for operations and maintenance, which brings the total bill to some $10 billion.

If you're wondering how the city would decide which projects to build first, the answer is a combination of "LOL, they're not going to tell you that" and "it depends on which projects the feds would support." The latter, of course, is how we got the streetcar.

Happily, there is a way out of this mess. The deadline for selecting projects is only imminent if the city insists on holding the referendum this year. But the city could hold it next year, spending the next 12 months working with MARTA on a prioritized list of projects based on actual estimates of ridership. Even better if the list represents an effort to build a coherent network that would boost ridership by more than the sum of its parts.

The only thing not served by waiting is our impatience. That’s the wrong customer to please.