How Do You Tell?

I had another good conversation yesterday with Prof. Sally Fincher, whose work was a big influence on Teaching Tech Together. The subject was notional machines, and in particular how to figure out as quickly as possible what someone’s mental model of program execution includes. For example:

After the following code runs:

A = 1
B = A
A = 99

what is the value of B?

Someone who thinks in Excel will (quite fairly) say “99”, because the whole point of saying B = A is to make B equal to A, while someone whose thinking is shaped by Java, Python, or the like will say that B is still 1.

Here’s another:

In what order are the words “before”, “during”, and “after” printed when this JavaScript runs?

const text = fs.readFile('data.txt', (err, data) => {

The natural answer (for some very biased value of “natural”) is before-during-after, but in Callback Country, the actual order is before-after-during.

And finally:

In what order are “F” and “G” printed when this code runs in R?

f <- function(x) {
  return (2*x)

g <- function(x) {
  return (2*x)


The answer in most languages is F-G, but the answer in R is G-F because of lazy evaluation.

If you have short questions that you use to probe or classify people’s mental models of program execution, I’d enjoy hearing about them: please give me a shout.

Tavish Armstrong sent this:

When a function A is defined, and the result of a call to a function B is assigned to one of A’s parameters as its default value, many people expect B to be called each time A is called rather than just once ag definition time:

def f(

f() # prints 2019-09-05 15:47:20.976051
f() # also prints 2019-09-05 15:47:20.976051
f() # and again...

Possible causes are:

  1. Nested function definitions like f(g()) do call the inner function each time they’re run.
  2. Many people (including some quite experienced programmers) don’t think of function definition as a thing programs do that is qualitatively the same as adding numbers.

In the wake of posts about Shopify's support for white nationalists and DataCamp's attempts to cover up sexual harassment
I have had to disable comments on this blog. Please email me if you'd like to get in touch.