The web, as I first saw it in 1996, was a truly incredible new phenomenon. Back then, on first seeing it through a browser, you instinctively knew there was something very special about this new medium. It was not so much a technical marvel as it was a social marvel. After all, the internet had been around for a while: only now was it beginning to be widely used. Let's recall some prevalent observations made to explain how special this web was:
These were so called mainstream observations. You would likely find them on paper--in paperbacks, newspapers and magazines describing in laymen's terms what this new thing, the web, was. A historical account of how the web had come to be, followed by conclusions comparing and contrasting the web with the older media.
If you read what the web was saying about itself, however, the emphasis was on its collaborativeness. “People of Earth ... A powerful global conversation has begun,” remarked the authors of The Cluetrain Manifesto. Because it was fundamentally a 2-way medium, the web was a dialog. You could write as well as read. Your voice mattered.
This was a collaborative medium. It was collaborative because it provided a facility, the hyperlink, through which every nugget of published content could reference another nugget authored and published by someone else. To date, it was the most efficient, easy-to-use, easy-to-make cross referencing means human kind had devised.
Knowledge is discursive. It is not produced in isolation; rather, it is manufactured in a juxtapositon of the new against a backdrop of the old. It is social. It is a production. An exercise in drawing on, referencing, previous works, rehashing, refactoring the past, sharpening the message, weaving a new story using older stories. And the web, together with its hyperlinks, is the latest manifestation of the this cross referencing activity. It is discourse magnified, if not yet in depth, then in breadth; if not yet in quality, then in quantity.
I believe it is the web's links that make it so special. Location-transparency is good, but overrated. We were well on our way there before the web came along: we had the telephone, and the fax machine, and the carriers were promising us video conferencing and something they were calling the information super-highway -- heck, we even had the internet, email and CORBA! It was only a question of time when we would all be location-transparent.
If web pages are cyber real estate, then hyperlinks are its cyber streets. The web is not just about cheap real estate; it is also about cheap roads. The interesting thing about the web is that when you publish something on it, someone else can link to it. When users reach your published work through someone else's web page, they interpret your work in the context of the other page that linked to it. Similarly, when your work references another web page, readers interpret the other page from the context set up in yours. In a sense, your page extends the page it is linking to. This extension of a web page can manifest itself in very many forms. For example, it may be complimentary or derisive. Or in a cocitation scenario, e.g. if it is linked from The Open Diectory Project dmoz.org, it may group the web page with others in a similar category. Either way, linking allows a web page to be put to new uses that the original author never envisioned. It weaves a nugget of old information into a nugget of new information.
The role of Ila is to compliment the existing hyperlinking mechanism. You could call Ila an existence engine. It provides a global view of all web resources it knows about and how those resources are link-related to other resources. This link-relationship can be explored in both directions: forward, from a resource that references another resource (e.g. a web page containing a link to another URL), and backward, from a resource referenced by another resource (e.g. from an email URL to a web page that contains a link to it). That is, if a hyperlink is a relation, then Ila allows a user to explore its inverse relation. Ila makes it possible to navigate links both in the forward and backward directions. A user endowed with this ability is link aware.
This deceptively limited feature set makes it possible to organize, publish and search information on the web using a purely link-based semantic. It allows anyone to annotate any given URL (more generally a URI) with new information. It allows directory structures to be built from the bottom up. It allows one to index data without having to put it into another database. It allows any one to extend and make available new data to users of the old data. It allows a decentralized, egalitarian approach for sharing information with other link aware users.