don't be a lurker!

thanks for stopping by, please donate your two cents to this post.

quotes on marketing & culture

disambiguity

December 16th, 2008 • posts i've written

When we create new things, we use convention to confront the unfamiliar. Clay Shirky has said that stuff doesn’t get socially interesting until it becomes technologically boring. Anything that relies on network effects needs convention to spread. We all have an email account; It’s the simplest and most abundant web technology. Online mirrors offline. If you want to write me, you need to know my address and you should include your signature. Our files, dispersed across the magnetic surface of a platter, are accessed by recalling folders and physical locations inside an artificial desktop. I spend my day at a desk, staring into a desktop.

The way we browse and participate on the web is still very much like the offline world. We follow paths based on landmarks. The web was built with human conventions, so that humans could interact on a granular level. Data was formatted so we, and not machines, could read it. What I’m talking about is the web of today vs the semantic web.

Tim Berners-Lee described the evolution as something like this… first we realized the key wasn’t the cable, it was the computers. Then we realized it wasn’t the computer, it was the document. Then we realized it wasn’t the document, it’s what the document was about.

The power and the chaos of the web is due to its decentralized nature. There is no backbone or tree-structure to hold data in place, hence the need for sophisticated search engines like Google and the existence of error #404, connections that have been lost or corrupted. Even the social web we have today exists on the same level as the documents, and not above it. Your social graph, or the giant global graph (the GGG, not the WWW) gets at expressing these connections in a more meaningful way.

The semantic web would allow data to be connected with other data to create meaning. Data mashups, using Google Maps to show incidence of crime across a city, mass transit, or even Flickr photos, begin to scratch the surface of what’s to come.

The cause of the hinderance is the lack of a common applied langauge. As decentralized systems grow, ambiguity creeps in. Human language grows more robust by coopting existing words and giving them new meaning (bark is what a dog says and what a tree has). Anything that is built to parse the data on the web must understand the relationships between that data to be effective. XML and RDF were created to be, together, the lengua franca of the semantic web. XML is a structure to hold data, and RDF is a three argument based system to parse the relationship of that data.

As we begin to understand the complex relationship of language, our search engines should be more successful at finding exactly what we’re looking for. As search improves, we’ll begin to see ‘agent’ bits of software that actively pursue connections between our data for us, so we can schedule a lunch with a friend by seamlessly vetting both of our calendars against open reservations at restaurants between us. Our devices will join the fray as well. Imagine watching a film on your set-top box and the soundtrack immediately downloading to your music device while your social graph tilts to connect you with other fans of the show, and additional content you’re likely to enjoy based on those connections.

sources:

http://www.sciam.com/article.cfm?id=the-semantic-web

http://dig.csail.mit.edu/breadcrumbs/node/215