Do you think asking a candidate to implement GOL in a 45 minute live coding exercise job interview and then expecting a fully working implementation if it’s clear they have never come across the problem before?
Devs I ask this come down 50:50 on if it’s reasonable or not.
And what are these hypothetical candidates interviewing for, exactly?
In the '80s and '90s, every time I got a new computer, one of the first things I'd do is write a very simple CGoL simulator, and then sit back and marvel at how much faster it was than the previous PC was.
So personally I guess I would have aced that question, in any number of languages and platforms... and yet I'm not really a particularly good programmer. It took me quite a few new-computer cycles before I started wandering down the various optimization rabbit-holes --
-- and I never got anywhere near as far as either HashLife or QuickLife. Now I just happily use other people's nicely optimized code, for the most part.
So... it's a problem that a sufficiently nerdy programmer type of a certain age will be very likely to have encountered before, and you'd learn completely different things about a candidate depending on the level of that past experience.
Devs I ask this come down 50:50 on if it’s reasonable or not.