In Hofstadter’s Surfaces and Essences, he and his co-author include a chapter filled with idioms, snippets of knowledge ingrained into a culture in a set of words that none other can perfectly replace, including the phrase, “once bitten twice shy.”
Having been forced to backtrack on my first day of biking to campus for my summer internship by a removed sidewalk, I admitted defeat and chose the next day a different route. I abandoned Dijkstra, Floyd, and Warshall, whose algorithms for calculating the shortest path are taught in lower level computer science courses, and took up a most reliable path mantra instead.
That is, I wanted to select the path that would most reduce my chances of needing to backtrack due to summer construction obstructing my way forward. This path, when interrupted, should have taken me and my rusty bike down a sidewalk that provides “back-up plans” whenever the original route failed.
However, each of these potential back-up routes must provide back-up-back-up routes, which in turn need back-up-back-up-back-up routes, and so on and so forth. It was as though I was looking for the “most central” route.
With this in mind, I biked from my house a mile uphill to the edge of campus, caught my breath at the crosswalk, and took not the direct, isolated path I had the day before, but instead the densely connected sidewalks situated at the heart of class buildings. These had, no doubt, come about through a century of paving “desire paths,” or trails through the grass carved out by heavy pedestrian traffic. One such sidewalk, providing a bypass to an already central intersection a few yards away, was introduced during my stay at WKU. Out of protest of its unnecessariness, a co-worker at the Kentucky Mesonet refuses to step foot on the newly laid cement.
I successfully reached my destination without backtracking, the goal of my most reliable path strategy, although I had been forced to alter my route due to a progression of color guard cutting across to the Fine Arts Center green space for practice. Imagine throughout my commute the second day the familiar GPS navigation system voice, the annoying, “Recalculating,” piercing the silence by insisting that I turn my bike or car or 18-wheeler around and take the most fuel efficient path—the shortest path to which I’d said, “once bitten twice shy.”
To figure the most reliable or most central path through WKU’s sidewalk network was trivial. All I required was a detailed map and a stroke of intuition. But how could I—now “we” as Dr. Atici and I have accepted this project as my Master’s Thesis—formalize this? Echoing Box’s quote that “all models are wrong, some are useful,” what variables should we include?
What we’ve come to, over the past two years of intermittent thought, is a pair of networks, one through which material moves and a second through which information travels about the first. Our problem is, though, that it might not be solvable that way, no matter how simply it reflects reality. ∎