Now blogging at diego's weblog. See you over there!

geography, software, and context

While I've been sort of away from blogging (and keeping up with blogs as well, I'm now beginning to catch up) as always I've spent time running around looking at the new apps and ideas that are thrown out there all the time. Of these, I've spent a fair amount of time thinking about mapping applications, for one reason: they puzzled me.

All that follows is probably obvious to many people, but hey, I won't mind that much for arriving late at the party and just be happy I showed up at all. :)

The puzzle was the sudden explosion of use and why it had caught me blindsided. Blindsided, in the sense that they'd be so popular. I've been looking at geolocation apps for a while now (the wireless ad hoc research community, where I spent some time during my thesis research, has been babbling about geoloc for quite a while now). But it had never caught my attention all that much. Partially the blinders that come from focusing on one topic are responsible, but on the other hand I found it hard to see them as more than niche applications. GPS: useful if you're traveling but most of us don't spend our lives in the car or train, and usually quickly develop enough knowledge of our transportation paths that a GPS can become redundant quickly when a route is coupled with habit. This is less true in the US, given that there's higher mobility and greater distance between places, but still you can't see using geolocation as something more than something that would be use ocassionally at best. And so on.

Geolocation is cool, I thought, but it's not massive, as say an RJ-11 plug is massive or as the Internet is massive. True, there is value is niches, and we could start yet another socratic discourse on the long tail, but...

When I started thinking about this, I began by questioning whether these apps where actually niche apps. In the end I realized that the apps are niche apps, but not just any kind: they connect meatspace with cyberspace.

That is, they provide context. Many people (me included) enjoy the geography of the virtual. We can see the towering structures, or as Gibson put it on Neuromancer, "lines of light ranged in the non-space of the mind, clusters and constellations of data. Like city lights, receding..." But that's only one half of the story.

If there's anything truly unique to pervasive computing, sensor networks, clothing with embedded circuitry, and a million other things that are going on right now, it is that they seek to provide a seamless bridge between the real and the virtual. Not a one-way connection mind you, but bidirectional. The ability to interlock devices with what's around us in a way that creates something that neither alone can provide. Example: object geotagging. Leave a marker using a webservice, automatically labeled with your Long/Lat, and others that walk by can find it. Have your clothes keep track of your movements and of who you meet, and let your shirt buzz softly against your skin when it realizes that you have an appointment in 30 minutes across town and you are still at a cafe, sitting still, with someone you know nearby ("someone you know" being defined in quick and dirty fashion as the same marker showing up in your travel history for the last 6 months, which makes it likely you won't just get up and leave that easily).

It's all about context, and it's not just geography--geography is just what we've suddenly got a critical mass of data on and the APIs to go with it). Geolocation is the tip of the iceberg. In itself, it's interesting, yes, and useful to varying degrees depending on your habits. But it's way more interesting for what it portends: finally, the emergence of real-world applications of context to software, and of software to the real world.

Categories: technology
Posted by diego on August 7 2005 at 5:09 PM

Copyright © Diego Doval 2002-2011.
Powered by
Movable Type 4.37