Opportunity marks the (parking) spot

I grew up in San Francisco. Car parking was a skill we bragged about at the dinner table. I have early memories of my mother (a native of Manhattan) showing off her car parking skills. She could wedge our Galaxie 500 into a spot barely inches longer than the car itself--and sometimes shorter if one of the cars on either end didn't have its brakes set. She also had, and still has, an uncanny knack for picking out the one person walking on a sidewalk who is headed to a parked car.

I know from parking.

And I knew, or thought I knew, that the Web bubble-icious company Parking Auction, covered in Wired last week, was a non-starter. The idea of this start-up: When you're about to leave a parking place, you can sell information about your imminent departure using a mobile Web site. If someone wants that information, they can bid on it. If they win the bid, they get directions to the spot that is about to be vacated and the transaction closes.

I thought it wouldn't work because there was a high-enough likelihood that before an arranged rendezvous between arriving and departing cars occurred, someone not in the system would swoop in and grab the spot, spoiling the transaction and inciting an especially nasty kind of road rage. "That's my spot! I paid for it!"

You can't sell what you don't own. Free and metered parking spaces are public goods. Parking Auction is an information broker, and it doesn't, and can't, control the real estate that the information it brokers is about.

Still, Parking Found co-founder Brian Rosetti feels that the business is sound. Orbiting for parking spaces is frustrating and wastes time and fuel. "It's just luck," Rosetti told me. "So I'm trying to create an incentive for people who have spots, to tell others when they're leaving." Rosetti is launching his service in Manhattan. The default price for a spot is $5. When I interviewed him on Friday, he had just launched and had yet to close any transactions. But he's convinced, of course, that it'll work. "You already see people getting in fights in New York," he says, and this system could reduce them.

My mother's take? "It'll never work. No New Yorker is going to sit in his car waiting for someone to be interested in his parking spot so he can make $5. They're in too much of a hurry."

Personally, I think making this a paid system that runs transactions between people is the mistake. It reduces the number of transactors in the system, and while it adds theoretical market efficiency, the Parking Auction marketplace requires not just buyers and sellers, but matching them to each other. That's a tough market to get going.

Related links
• Tiny start-up tackles big driving hassle: Parking
• Reporters' Roundtable: Where is navigation going?

So I propose a totally free system that pools data. Instead of making money, you get rewarded for putting parking information into the system with, get this, credits for information about actual parking spots. The motivation to put spot data into the system would be clear: Parking in, parking out. You could even limit information for low contributors. More importantly, users could be rewarded for putting spots into the system even if there were occasionally no "buyers" for their specific spots. (Verifying the accuracy of their data is a problem I haven't solved.)

A free system would also make finding a spot a one-step process: I'm here, show me nearby spots. The auction system requires extra steps: Are there spots near me? Yes? How much? OK, I'll buy. Did I get it? Yes? OK, where is it? Now, is it still there?

There is a problem with my plan: It's already out there. A small company called Google has an Android app called Open Spot that is supposed to do exactly what I just proposed. But it's apparently a forgotten service. I tried it on Friday and it located zero spots for me in San Francisco. (Someone try it in New York and let me know if it works.) I think limiting the app to Android devices is a problem; also, I think there are user interface challenges for this data market that are unsolved.

Street parking data is so time-sensitive that a market for it--free or paid--won't work until the information in the system is more complete, and both data entry and query are easy, safe, and lightning-fast on mobile devices. I think this idea can work, but the current apps I've seen are either too economically limited or too slow to tackle this interesting and dynamic information market.