Now, how is this supposed to work exactly? Take the quick performance and terrific usability of Google maps, take the science of GPS enabled fleet management gear already installed in taxi cabs and van services, and you have Google's Ride Finder. The result, for a select few metro regions in the US, is a Google map with plots that represent individual taxis or vans.

Not sure why I would want to know where these vehicles are, unless I'm the dispatcher. In fact, that's just the capability that guys like Mobile Knowledge sells. If I want to hop on a Super Shuttle van to the airport, do I go to the site, find the nearest one, and chase him down the street to let me on? Not unless they plan on booking these vans in a different way (and I don't think they are supposed to take on un-scheduled pick-ups, no?). Maybe I can just call the dispatcher and whine "Why should I wait, he's three blocks away from me now??" Doesn't sound like a great deal for all the Louie DePalma's of the world.

It's a neat feature, and showcases the power of GPS technology and the use of the web. I can dig that (see my own experiences on this topic). And using the existing base of GPS enabled fleets makes practical sense too, for now. But my silly, yet apt description above makes me wonder where this is going. Unless Google is looking to replace fleet dispatch with some sort of free-ranging online click-and-summon travelling salesman algorithm, I don't see the value yet.

Use of GPS fleet tracking to this point has been limited to managing workers in the field. How about deploying it as an added-value consumer-facing service? Surely everyone else has waited for a delivery that required a signature, only to be out of the house (or in my case, probably on the toilet) when the doorbell rings. Chasing the UPS or FedEx guy down the street is kinda fun, sure, but it would be nice to get a SMS or email saying the guy is in your neighborhood -- or that your segment of his route is coming up next. Of course, why not do this for all home or office deliveries? If the infrastructure is already there -- why not extend it to be a value-added feature for the consumer, and a competitive advantage?

Here's another possiblity: imagine if this Ride Finder interface was actually a map, not of cars that are taking people to different places, but of positions of "open-ended delivery" people. These guys are all around the city, and will auction off their time to do your bidding. Example: I see on the map that someone is near a store or restaurant that's a bit too far away from me to be convenient -- I send out a request, a price is brokered for any additional fees, and exactly what I want is on it's way. Include in specifics for a product via passing the data from a handheld UPC scanner, and you can guarantee you get what you want.

It's like the glory days of Urban Fetch or Kozmo, but without inventory or infrastructure. And there are, of course, hurdles to solve -- does the delivery guy have to front the cash? I guess you'd have to have registered users, and you'd charge their credit card before you accept the job (if you knew the price, that is).

So, why not just do this with a regular, perhaps high-end bike messenger service? Because I think the service would work better if the delivery guy's location is public knowledge -- sure, someone who's now in Queens might claim he can take my gig and get it to me in half-and-hour, but I don't believe him, and choose to book with another guy instead. Count in my skepticism the numerous times I've been assured by the car service dispatcher the guy is 5 minutes away, when you know he's 30 minutes away and stuck in traffic with another customer (been on that side of it too, including driving up a sidewalk as the dispatcher yelled at him for lying). More direct contact to the actual individual delivery guy is good, so you can have more trust in what you're ordering and moving about (rather than merely sealed up boxes or envelopes flapping out a messanger bag). You're paying more, and asking more of these delivery guys, so you need more trust there.

Ok, so that's my thought on the matter. Had sketched this to prototype off the new Nextel Blackberry GPS units, but don't have the time to execute it. That's just one of my schemes to replace fleet dispatch with some sort of free-ranging online click-and-summon travelling salesman algorithm.

Now who's gonna give my mom and dad a ride to the bowling alley? That's what we gotta figure out.