Page:Aaron Swartz s A Programmable Web An Unfinished Work.pdf/63

Rh

Well, we've been through a lot together, you and I. We've built up an application from its humble URLs to its high-falutin' notions of democracy. Along the way, we've made its world safe for robots, query systems, researchers, and Richard Stallman. But how do we get from this kind of website to the grand vision of the Semantic Web we've heard so much about?

Let's start by realizing that just being on the Web is an amazing thing. URLs provide a uniﬁed addressing scheme for any document—a pretty miraculous thing. Imagine trying to explain to folks of yesteryear about these incredible words which you could give to anyone and they could take it home, punch it into a box they have on their desk, and get just the article, picture, or video you meant. To a generation whose idea of document retrieval is driving to the local library, ﬁlling out some forms, and waiting a few weeks while they tried to make sense of your subscription and hunt it down, this is a pretty serious change.

But REST took it even further. By making the documents accessible by search engines through a standard protocol, you no longer even need to know the right URL for the thing you want. Just type a few choice words into Google and boom! back comes just the thing you wanted in a quarter of a second.

Of course it's not just Google; REST makes possible an interconnected tapestry that supports everything from web browsers to web editors to intelligent translating intermediary proxies.

A hard act to follow. But then came the ability to get not just documents back from these far-off servers, but data. By importing and exporting raw data, we made it possible to switch software programs, providing a somewhat-free market of competition among products, creating massive consumer surplus. Go us!

But we didn't stop there. Just letting users take their data home with them is weak brew compared to sharing it with the wider world. (Picture a wide world of