Its hard not to get excited about the local search services recently announced by Microsoft and Google—and Amazon now is jumping into the mix. Merging street maps with aerial photography, point-to-point routing and a database of interesting places make for some impressive screen displays. Microsoft is calling its effort Virtual Earth. The demonstration I was given even included 45-degree aerial views, allowing users to see the sides of buildings as they move down a virtual street.
Theres no question: These things make great demoware.
What the companies are building are geographical user interfaces that display database information on top of local maps and aerial photography. It probably overdoes it to call them geographic information systems, but they are cut from that cloth. Maybe these services will become a poor persons GIS as more information is poured into the database and they gain functionality.
But making local search useful in real life—by people who actually live in the “local” area—is likely to prove much more difficult than building the demos or even creating the geographic user interfaces necessary to drive the services.
Heres a problem I posed to the Microsoft folks showing me their Virtual Earth demo. They had already shown how I could ask the service to find a steakhouse near a downtown Seattle hotel. They then dragged the map to another location—I guess I changed hotels—and it magically updated the restaurant list for the new location.
Lets forget for a moment that the first generation of the MSN virtual planet doesnt know about things like one-way streets, so its probably good I left downtown Seattle. But what the service also doesnt know, and probably wont for many years, are the answers to more complex questions.
For example: Ive just arrived at my hotel. Its 10 p.m., and I want to know about all the restaurants that are within walking distance of my hotel that are still open so I can have a late meal.
Now there are already plenty of databases of restaurants available online. But have you used one lately? I am not sure anyone does, usually because they are significantly out of date and not very encyclopedic. I have never seen one that allows such a time-of-day search, and I wonder if hours of operation are even a part of these databases.
So I do what everyone else does: Rather than go online for half-baked and half-useful information, I call downstairs and ask the concierge or doorman where to eat. And even if Microsofts Virtual Earth were up-and-running right now, thats still what Id most likely do.
Next Page: Improving quality and coverage.
Page 2
To improve the quality and coverage of its content, Microsoft says it plans to supplement (and correct) its databases by allowing users and businesses to enter their own information. Thats not a real solution to deal with the information glut local search services must manage.
In fact, given the mayhem that besets the Internet when someone can make a buck by ripping off someone else, I wonder if opening the database to outsiders entries will make the database more or less useful. Its not like it will be possible to have a human being verify the new data.
How long do you suppose it will be until Googles local information becomes as polluted as the rest of its search results? Googles biggest problem isnt Microsoft or the need for local search; its the ever-decreasing quality of its product.
Its hard for me to even imagine the amount of information needed to turn my city of 75,000 sun-baked Californians into useful part of a virtual world. Sure, you can limit the level of detail and coverage to the sort of touristy and Yellow Pages information already widely available online. If that becomes the came, then all Microsoft and Google seem to offer are what may prove to be a better user interface and experience.
On that score, aerial photos are impressive when you first see them, but even with a map overlay they can be hard to use. I have access to a lot of aerials, but only use them on special occasions when I really need to know whats at and around my destination. Trust me, the “wow” factor of overhead views wears down quickly.
But the biggest problem with local search is making it portable. Sure, the new services will all look great on your desktop in the office or at home. But how well will they work on a mobile device?
Screen real estate on a handheld anything is going to range from small to teensy, depending on the specific device. Current user interfaces, even on the best ones, dont make you want to spend a lot of time searching for things, and even if you did, the download times for useful amounts of data can be fairly long.
As someone whos never met a map he didnt like, Im prone to get excited about these services. But I also think about all the ways Id like to use them and how much information it would take for each application to be useful. Then I become concerned. Add the challenges of making these services useful on a cellular handsets tiny screen, and Im downright worried.
Still, search services are the new battleground. Google needs to do something to prove its worth $200+ a share, while Microsoft needs to find revenue sources beyond selling boxes of software. There is incentive and opportunity here, and companies will spend heavily trying to get this right.
They are likely to prevail, though it could take a very long time. As the work continues, I hope the competitors understand that to be successful local search needs to be more than a user interface gimmick that covers the flaws in a so-so collection of data.
Contributing editor David Coursey has spent two decades writing about hardware, software and communications for business customers.