Base 10
What does it mean to have a canon? In English studies this is usually a body of texts that it’s assumed most serious scholars have read deeply and which somehow embody whatever characteristics or themes are deemed most relevant to a given perspective (e.g., a Western canon, a Shakespearean canon, a post-colonial literature canon, and so on). In other words, the members of a canon act as pillars in the foundation of a shared body of knowledge.
In physics, we don’t really have a canon. There are many famous historical papers and a few books and textbooks, but mandatory deep study and a shared list of “why these are canonical” (even if hotly debated) is not really in our culture. There are perhaps, to a degree, canonical problems—physics problems everyone sees and attempts (recognizing that, ironically, who your “everyone” is will vary by sub-field). These are most often presented in one of two groupings: by subject (mechanics, thermodynamics, astrophysics, etc.) or by math (differential equations, group theory, etc.). Only ever so rarely are these problems grouped by core concept in any consistent way (perhaps Feynman’s three volume lecture series is the best example here).
My roving imagination and mind were hard at work again when I came across a piece about speed reading. What captured my attention most was the emphasis to (1) first learn the technique, then (2) practice the technique for speed ignoring comprehension, then (3) practice the technique at speed with comprehension. For a while, it has seemed to me that analogical thinking is a good test case for a discovery strategy applied to active, professional research. But how to do that?
I have some ideas for how to synthesize a few operational analogical processes, which I’m hoping to work on with the help of master’s students this Fall semester. But the speed reading piece reminded me that practice is key. So how to practice? Well, in English studies you practice critical thinking skills on the canon where you can compare your results with others, then you venture out into other non-canonical areas. In physics our own canon is problems, so that means that to study and practice discovery strategies one will need a good discovery canon. I’ve nicknamed the physics discovery canon I’m developing “Base 10.”
In my experience as an undergraduate student I always followed what I called “The Rule of 10”: practice a new math technique ten times before applying it to what you actually want to solve. This was a necessary expedient since, by the time I started back in on my physics degree, it had been 5 or 6 years since I had studied the subject and I took the minimum number of courses (which meant little math) to get out of undergraduate and on to graduate school as quickly as possible (a money problem, not a time problem).
But of course, this rule of ten strategy also requires problems to practice on. Hence, base 10 as a general rule for the number of test cases I need to try something out. Now my natural inclination toward favoring analogical discovery strategies over others, combined with another math-inclined strategy known as “easy cases” (aka “toy models” where you keep the simple stuff and leave out the complicated details) has led me to believe that the standard groupings of physics problems may not be suited to my needs. I need more conceptually useful categories right now, not categories that are mathematically similar or topic dependent. It’s just a hunch, but worth an attempt.
So, I am slowly compiling my base 10 physics discovery canon to practice discovery strategies on. The worst that happens is a little trial and error (technically, another discovery strategy which goes by the formal name of “generate and test”). And if it doesn’t work out then, as I always tell my students, there’s a reason why it isn’t called “trial and success.”