Context
Recently, I connected with James Burke and his KnowledgeWeb project. This project seeks to overcome the blindsided pursuit of specialization that loses credibility because it lacks context. The idea of tapping into a database of research and intelligent connections is straightforward, but the execution is almost inconceivable. I envision this like a cultural databank, a memory bank of the human race, but instead of happening, it is created. It must be mind-boggling to create a model of part of the mind that works as a partner with knowledge workers. I first heard James Burke speak at the first Microsoft CD-ROM conference. When I met him, almost 20 years ago, I didn’t understand what he was talking about, but I didn’t know it. Now at least I know it.
Back then I was an expert in an industry so new I had to start my own company to work in it. I began to compile a database of CD-ROM sales. For many years, I knew everyone who sold anything on CD-ROM, as well as everyone who made the discs and drives. That changed when CD-ROM intersected with the advent of A/V-capable personal computers, GUIs, search software and digitized data, but I kept on top of the sales, building the data until the mid-1990’s.
Prudently, I thought, I had a policy of projecting forward as many years as I had historical data. Then the multimedia market peaked and contracted, leading to a commoditization of discs and drives that has followed it through the transition to DVD, and a psychological devaluation of packaged content that has followed it through it’s migration to the internet. I had not predicted it, but, in retrospect, I knew why. I was looking only at my data on CD-ROM content and technology sales to make my forecasts, talking only to those vendors, thinking like a specialist. Multimedia CD-ROM maxed out because people did not have sufficient time or resources to keep growth going, without giving up existing linear forms of entertainment, like pay TV, home video, and music that is. It harkens back to what another early Microsoft CD-ROM Conference speaker, the late Gary Kildall, said, ‘the extent to which people want to interact with their televisions is unknown’. We discovered the extent in 1995, and it was a lot less than we thought.
As we began to prepare to forecast DVD, this time two years before commercialization with no historical precedent, we built a crude contextual model of video and online revenues using OPD*. I became so fascinated with the interdependencies between the various formats, we abandoned primary data collection for DVD after a few years when everybody else started piling on, and began work on a model of all consumer entertainment spending. Throughout the Internet boom and bust, when others were creating fantastic forecasts to support the fantastic investments, we had our heads down collecting and crunching numbers, and escaped the blame for the resulting crash. The result (or rather, the work in progress) is the first and only comprehensive, worldwide, zero sum spending model for consumer entertainment content and electronics, the InfoTech Forecast® 2020. Spending for content and technology is capped by the estimated disposable income allocation for home entertainment plus the estimated ad spending for content sponsorship and all forms of spending are accounted for, even if we have to interpolate them from other known extents and rates of change. The model is applicable to many phenomena, of course, not just consumer spending, but we had to start somewhere, so we started with a market whose vendors struggle to get their heads around it and fling extraordinary sums at to get their teeth into it (and hopefully some at us for illuminating it).
Clearly this is what the KnowledgeWeb wants to be, a classic application of a contextual model born of common sense, aka prior mistakes. I wholeheartedly endorse this effort because in my own experience being able to look at something in context is essential to understanding its behavior and fate.
posted by julia b schwerin