Better Practice, Dec. 2018

42 \ December 2018 \ http://www.phparch.com

Education Station

Interview Coding Challenges

Edward Barnard

I’ve encountered job interview coding challenges over the past year. In each case, the company

or recruiter said people have trouble with these. Let’s talk about that! We’ll look back at the

Education Station theme for 2018, pick a new theme for 2019, and tie it all together.

When arguments begin over which

is the best programming language, the

standard PHP joke runs like this:

Question: What do you call a good

PHP developer? Answer: Employed.

This joke does explain the general

perspective for PHP people. We’re

pragmatic. We deal less with theory

and more with practice. We translate

needs and requirements into something

concrete—a web application.

Generally speaking, we create visual-

izations—things which can be seen or

used—rather than algorithms.

Meanwhile, the days where employ-

ees stay with the same company 20-30

years are long gone. We move around

or move on from contract to contract.

For many of us, this means formal job

interviews. Many of those interviews

include coding challenges; that can be

a problem.

Let’s get to it.

Goals

Each of the challenges I encountered

was of the form “write the code to make

these tests pass.” There was a clear state-

ment of the problem and instructions

for running the tests. None of these

challenges had anything to do with

building web pages.

That fact seems strange to me. Why

bother with a coding challenge which

has nothing to do with the actual job

requirements?

Are these Kobayashi Maru^1 scenarios,

that is, no-win situations? Yes, and no.

One objective of the coding challenge is

1 Kobayashi Maru:
https://phpa.me/wikip-kobayashi-maru

to observe how you perform during the

scenario. It can be a win-win situation

rather than a no-win situation. Let’s see

what’s behind these challenges.

Fizz Buzz Test^2 explains the FizzBuzz

Test:

The “Fizz-Buzz test” is an interview

question designed to help filter

out the 99.5% of programming

job candidates who can’t seem to

program their way out of a wet

paper bag.

That 99.5 percentage comes from

the Coding Horror article Why Can’t

Programmers.. Program?^3 by Jeff

Atwood, co-founder of Stack Overflow:

I was incredulous when I read this

observation... “Like me, the author

is having trouble with the fact that

199 out of 200 applicants for every

programming job can’t write code

at all. I repeat: they can’t write any

code whatsoever.” The author... is

evidently turning away lots of

programmers who can’t write a

simple program: “After a fair bit of

trial and error I’ve discovered that

people who struggle to code don’t

just struggle on big problems, or

even smallish problems (i.e., write

an implementation of a linked list).

They struggle with tiny problems.”

Both of the above articles are

worth reading in their entirety (and

they’re short). The Coding Horror

comments section debates the merits

2 Fizz Buzz Test:

http://wiki.c2.com/?FizzBuzzTest

3 Why Can’t Programmers.. Program?:

https://phpa.me/codinghorror-cant-program

of FizzBuzz-style questions, e.g., does

not being able to implement recursion

mean you’re not a good programmer?

Coding Horror explains why inter-

viewers are presenting these sorts of

challenges—people “can’t seem to

program their way out of a wet paper

b a g .”

This claim is both harsh and unfair.

If 99.5 percent of job candidates have

trouble with these coding challenges,

that tells us:

• These coding challenges have noth-
  ing to do with and are irrelevant
  as predictors of actual job perfor-
  mance, or


• Our training system has nothing
  to do with and has no relevance to,
  our real-world jobs, or


• Both of the above.
  However, the coding challenges exist.
  We’ll look at my strategies below, but
  first let’s consider the explicitly evil
  cousin, a specific type of whiteboard
  interview.

White Board Interviews

Adrianne Jeffries in Programmers are

confessing their coding sins to protest a

broken job interview process^4 shows how

a particular form of whiteboard inter-

view selects against a diverse population.

This interview style, widely used

by major tech companies includ-

ing Google and Amazon, typically

pits candidates against a white-

board without access to reference

4 Programmers are confessing their coding

sins to protest a broken job interview process:

https://theoutline.com/post/1166/