760: on berners-lee’s information management proposal

I read Tim Berners-Lee’s Weaving the Web while I was working on Polaris. We had been using the old WinHelp spec, but it’s .rtf format was proving restrictive as Polaris matured and more functionality was added. I’d used hypertext before for help systems (there was that clunky HyperCard system for Seagull Press in ’94), but I felt like I needed to understand hypertext in a broader context. That’s why I picked up the book. I read it in a few days and went on an evangelizing kick, trying to get all of the writers and testers in our little group to bone up on hypertext theory. I argued that the morphing HTML specification and more scalable help systems (such as Compressed HTML Help and JavaHelp) would require us to understand what we were creating. The old practice of committing text and images to the ether and wishing for the best wasn't going to cut it any longer.

Looking back, it’s obvious I really didn’t get it myself. It seems we would have been better off by simply immersing ourselves in a hypertext environment, as Berners-Lee described in his CERN proposal, and let the system take a shapes that reflected our changing needs, skills sets, knowledge, etc.

There are two aspects of Berner-Lee’s original proposal that are only now beginning to fully emerge on the web. The first is the ability to modify (edit) information directly through the browser. The WorldWideWeb browser (and later Opera and Amaya) would provide this feature. And in fact, a number of knowledge management systems, such as Canterbury, would ship with propriety viewer/editors. But I think it was the distributed nature of the connectedness of the Internet (and later the web) that made (and continue to make) this feature impractical outside of dedicated communal knowledge systems, such as wikis.

The second aspect of the proposal that took a while to take shape is what we identify today as Web 2.0 technologies. Before mash-ups, RSS, XML, and new media, technical communicators were struggling to adapt the first HTML specifications to create interactive help systems. The closest we came was context-sensitive help (and later embedded help), which allowed us to push specific topics to the end-user at specific points within a workflow. It was clunky, but it worked.

Reading the CERN proposal brings a lot of that work into focus in a good way – a fun way.

More specific to how we’re being asked to consider the proposal against the rest of this week’s readings, Berners-Lee is imagining a system dramatically different from the one-dimensional “fetch-retrieve” system envisioned by Wells and manifest in the early computing systems of the late 1940s and early 1950s: “For this to be possible, the method of storage must not place its own restraints on the information. This is why a ‘web’ of notes with links (like references) between them is far more useful than a fixed hierarchical system.”

In specifically addressing the problem with “tree” hierarchies of data, Berners-Lee is re-articulating Bush’s vision of non-linear, dynamic relationships among information objects. Perhaps more important to the progress narrative among the readings, Berners-Lee is refining Licklider and Taylor’s imagining of a communication system with the capabilities to capture communicative models: “When describing a complex system, many people resort to diagrams with circles and arrows. Circles and arrows leave one free to describe the interrelationships between things in a way that tables, for example, do not. The system we need is like a diagram of circles and arrows, where circles and arrows can stand for anything” (Berners-Lee).

Berners-Lee’s vision is impressive when it’s considered against the modern web. Here’s a guy who was responding to the age-old problem of retaining organizational knowledge in a way that would be useful to those who stayed on and those who came in. His simple proposal, coupled with enabling technologies, has quite literally changed the way the world communicates.