Back in 1999, I thought of a neat idea. I’d been reading about the new generation of “location based” phones, which were going to be able to pinpoint where they were located, down to a few feet.

This, I figured, would be an insanely revolutionary step. It would open up a whole new world of data. We could start implanting information in physical locations — for example, leaving a message for other people to stumble upon as they walk down the street. Imagine walking down the street with your phone set up to sniff out messages that other people have left. You find a notice in front of a restaurant, where someone has warned you that the place sucks; you read a short memory someone has of a streetcorner where they first met their lover; you read political “tagging” of parts of the city by activists. I called this idea “spatial messaging” (which is a nastily un-catchy name, but whatever). With spatial messaging, you’d use your body as the browser — surfing through a physical world loaded with information.

The more I thought about it, the more I realized just how profoundly this would shift the stakes of the Net. After all, the whole social power of cyberspace was that it decoupled information from the physical world. The real world is shaped like a map; if I’m over here in city A, I’m going to hang out with people who are physically in or near A. In contrast, the Net is organized not like a map but a library. That’s why the Yahoo hierarchy has a Dewey-decimal-like organization. The Net allowed all the model-train freaks, the weekend nudists, and the Star Trek nuts to all find each other no matter where they were physically.

Spatial messaging would completely reverse those stakes. When you implant information in the physical world, geography becomes the most important organizing principle. What precisely would that do? I truly had no idea, and still don’t; I’m assuming that when this finally happens — and it will — the results will be as unpredictable and weird as the Net. When I interviewed Vint Cerf a while back, he said he had no clue that the Internet protocols he wrote back in the 60s would eventually result in Napster, chain email, and porn sites.

At the time, a group of geek friends and I actually tried to create a spatial-messaging application. But back in 1999, the phone companies hadn’t yet gone far enough in bringing location-based tech to reality; we didn’t have any baseline network on which to program. So instead, my friends and I set up Beaker.net, a web site that functioned like a Geocities for cell phones: It would let you quickly produce a little teensy site that could be viewed on mobile devices, even if you didn’t know WML coding. (We built the site in our spare time as a research project, and somehow — without any advertising — got about 10,000 users, which was fun! But it died after our free hosting service accidentally erased it last year; we’d all moved on to new projects, and hadn’t really maintained it. You can see a snapshot of the home page here, if you want, on the Internet Archive.)

The point is, with Beaker.net, we were trying to create a tool that let everyday people produce kooky, weird, offbeat stuff for mobile devices. The phone companies were completely ignoring this area. In early 2000, the only “content” they syndicated for those godawful WAP-browser phones were sports scores, horoscopes, and stock quotes.

But as far as I’m concerned, Internet applications really explode when the infinite monkeys of the Internet — Joe Weirdos the world over — can generate crazy stuff that cracks you up and gets you to forward it to a zillion friends: Populist, popular culture. If we’d waited for mainstream companies to fill the Internet with content, we’d have about 300 web sites in total right now. So I figured that the mobile Internet needed the same shot in the arm. (As it turns out, by late 2002, many mobile devices were able to read full HTML, so the need for mobile-phone-specific sites thankfully faded. And, interestingly, the biggest “content” revenues for the mobile industry were coming from downloaded ringtones and phone gaming — proving once again that culture is the driver of new technology, and not freakin’ sports scores and stock quotes.)

But … location-based technologies now need the same cultural impetus. At the moment, mobile companies are obsessed with using location-based tech to sell the same dull, boring, or horribly invasive information to you. They’re working on technologies to blast Gap ads at you when you’re near a Gap; or, like the the Pinpoint company , they’re setting up location-based tools to to track the location of employees for their bosses. Nice.

The point is, in pretty much every case I’ve seen, mobile corporations are only developing one-to-many applications. Some of these will be very useful, of course — such as databases that help you instantly locate the nearest subway, or, Vindigo-like, a good Italian restaurant. But it’s all about having big corporations implant the physical world with data.

But, like I said, Net applications only truly explode culturally when they’re a many-to-many phenomenon. Think of the hugest trends on the Net; what are they? Instant-messaging, blogging, email, chat rooms, and file-sharing. They’re all many-to-many; they all harness the desire people have to communicate with everyone else in weird, strange, and unpredicted ways.

Spatial messaging, or something like it, is much more likely to capture the imagination of mobile handset users than Gap ads, I’d say.

The good news — and the news item to which this incredibly bloated intro was pointing — is that some research parks are cluing in to this. In the new issue of Discover Magazine, Steven Johnson writes an extremely cool new column discussing the emergence of “GPS-based hypertext”, the very concept I was bandying around with my geek friends. Johnson starts off by pointing out how the GPS game of “geocaching” is a prototypical type of spatial link. You set up a stash of goods somewhere and leave nothing but the GPS co-ordinates as a clue for others to stumble upon:

The great breakthrough on the GPS horizon lies in thinking of those geographic coordinates as a real-world URL. In other words, think of those digits not simply as a description of a point in space but as a place to store information. Today you can create a Web address and publish pages and pages of anything you want there. But soon you’ll be able to take a GPS location—say, 40°43.833’ N, 073°59.814’ W, the coordinates for Washington Square Park in New York—and publish material there as well. Anyone walking through the park would then be able to browse through the data you’ve uploaded.

And as Johnson notes, the key thing is that this needs to be a system where anyone — not just the Gap — can publish spatial information:

“Instead of having just tourist information, the system would be open,” says Swedish researcher Fredrik Espinoza, cocreator of an experimental tool called GeoNotes. “There would be much more social activity.” Espinoza’s vision includes a filtering system for retrieving GeoNotes that have been posted by friends or other trusted sources, like the buddy list of Instant Messaging. Imagine, for instance, that you stumble across a beautiful side street in a historic district, the sort of urban discovery you might tell your friends about the next time you meet them for coffee. With GPS-based hypertext, you could leave a virtual note hanging near the street, addressed to your 30 closest friends. The next time they happened to stumble through the area, the text would pop up on their PDA screens: “Hey, come check this out…”

Here’s hoping this stuff comes along soon! The Net — and the physical world — will never be the same.