It's so stupid to spend months preparing for interviews just to learn something that you will never, ever use. There are libraries for a reason.
What you think is good and commendable to me views as a horrible practice in our industry.
That clearly says to potential players that the organization doesn't care about chess, and won't have the best chess players in the tournament. Instead, the only people playing are those willing to jump through hoops on command.
First, I agree bad interviews are bad.
That said, bad or un-necessarily difficult hiring processes dont automatically mean there is a better company out there.
IF there is, it isn't a given that the better company will hire you.
Similarly, it is not even a given the employees jump through hoops and get in are inferior to those that do. This is a numbers game, and can go many different ways.
On one hand, I appreciate the uncompromising nature of refusing to play the game, but on the other, I dont think that everyone is doing it with open eyes to the costs they are paying.
I have seen a few people progress from: 1) being uncompromising to 2) worse outcomes than if they compromised, to 3) bitterness and anger.
Back to the analogy: If your life dream is to play chess but you refuse to play poker to get in the door, you may end up bitter and playing with mud in the street.
I agree as a candidate it can be demoralizing, but thats the exchange usually - prove you can do the job (or something close to it) and you get hired for it. It's just that your job isn't street sweeping or generalized labor -- and it's helpful to know if you're going to be the type who: Couldn't write leftpad without a library, would depend on leftpad as a library, or would understand the implications of using a library like leftpad and rely solely on trusted functions like isEven instead.
Yes, you're a programmer, you will get libraries in your day job - but as your hiring manager, I don't have the time to tell you you won't have access to most of the ones you're familiar with because we're doing real-time or large-scale compute or javascript widgets on embedded devices - I just wanna see you do as asked.
If you decline - I've seen what I needed.
ChatGPT is a really good junior programmer who can write basic functions per best practices, and when it makes a mistake or does something weird, a senior SWE will spot it, and can fix it with much less mental effort than writing the function.
Job's are not a popularity contest - it's about if you'll do what's asked of you, or if you'll spend more time arguing about the right way to code something rather than solving the business need -- and for my purposes, I need the business problems solved, not someone who gets self-righteous when asked to show an example of them working when applying for, ya know, work.
You don't have time to explain the job requirements or stack?
Many cannot, and those that argue the premise don't even get a chance to try.
If you think you're being evaluated on your knowledge of the stdlib's sort(), you misunderstand the purpose. You're being tested not because you can hit compile and it passes, but you're being tested on if you're able to follow directions, work within scope, solve problems and explain your thinking. If your thinking stops at "I'd never do that", so does mine.
That applies to any interview test though, you don't need to contrive an absurd scenario to do it.
Tangent — how is isEven trusted? It seems people usually categorize leftpad and isEven pretty similarly (e.g. pointless bloat deps waiting to be next supply-chain attack), so I don’t follow the distinction you’re making.
As a rule of thumb (not quite _always_ true), if you couldn’t implement the lib yourself you will be showing some skills gap in some way or other