God wants your footnotes on the side, damn it!
By its end, I had a consciousness expanding experience. It is a true joy to attend a presentation on presentations from a master of presentation. The economy of thought and words he used to convey meaning was really astounding in retrospect. One main thread stood out during the six hours or so that I mutely absorbed sage advice, however, and that thread concerned the scaling of information.
It seems like scaling information to the point where it becomes a useful tool for generating conclusions and causal relationships is the true focal point in analytics. Tufte presented reams of stock quotes for a company over a massive period of time, rescaled into a single small squiggle representing the movement of the stock price, with four dots representing
• the starting value of the period
• the ending value of the period
• the maximum value of the period
• the minimum value of the period
In just glancing at the squiggle, the information had been scaled to an inviting size for human analysis. This seems to me to encompass the same function as asymptotic notation in computer science, scaling the behavior of the function (with respect to the input size) to an inviting size for comparison. I can only presuppose that if the fundamental property of analytics is scaling information so that conclusions can be inferred, then computers may be able to assist cognition in far more profound and serious ways than I had previously thought by simply auto-scaling the data to the appropriate perspective.
Professor Tufte must be making a <insert expletive here> storm of money from his seminar circuit to be able to purchase first edition copies of Galileo's works and erect large stainless steel sculptures in his backyard. I imagine such hobbies are quite costly indeed, even if he does claim to collect large scale scrap from his local nuclear facility.