knowledge based questions in interviews to check if you've internalized and memorized stuff feel awful

architecture and system design questions when you don't actually get to do any of it on the job also feel awful

i'm also a scrub

I talk about stuff I use. Like MVC. I have had great success to make programs very easy to maintain and automate. It also made it so I could upgrade the GUI without affecting the rest of the code base.

@YourMom

> It also made it so I could upgrade the GUI without affecting the rest of the code base.

I know that this is the prime example and it’s being taught in school/uni, but have you asked yourself what that actually means?

Upgrading the gui? Changing to another gui? When does it ever happen?

I was a frontend dev for decades and used a lot of different ui frameworks on different platforms.
I never needed to upgrade or change the gui. At least not in a way that it would benefit from an architecture like MVP.

It‘s still good to separate UI and business logic. But not because of that.
Some valid reasons are mocking and testability (unit tests)

Let me explain you tests. Tests have four possible outcome. True positive, true negative, false positive, false negative.

It means here:

True positive: Deserve job and get job.

True negative: Don't deserve nor get job.

False positive: Get job but be bad at it.

False negative: Don't get job, but would have been great at it.

If the candidate pool is big, companies can afford to optimize for low numbers of false positive. That's the expensive one. They only are willing to dare false negatives when they are in a bind and don't have any other choices.

All the "I don't like the practice of hiring, that is so unrealistic" overlook that there is a strong correlation between being knowledgeable at a field and good at it and misunderstand what companies are optimizing for. So, you don't get the company's needs.

@TrayKnots wow. That was interesting. But also sad, if that‘s true.

@Lensflare You know what's even sadder? You require considerable brain power to understand what's going on. But not to do it. Those companies haven't thought that through.

They tried something. Something worked. Companies that tried stuff that didn't work were out-competed and bought up or vanished. New companies mostly copy old ones, after all, they want to look professional. Which is code for copying other companies.

Which means, we have a population, in which only the best survive, then we have a new round of copying and mutating, and then again only the best survive.

It's an evolutionary algorithm. Like always, no real thought, mostly evolution. I'd feel better at it if they just were smart, but the lucky out-compete the smart by numbers.

@Lensflare I dunno, I was using pygtk and was thinking of switching to Qt. But it never happened.

edit: I did automate things though.

personally i havent exactly had any normal-ish interviews of any kind in my career... but i kinda assume you could use the same principle as i used in public speaking, debate, arguing with teachers, anyone trying to probe for random overly specific facts instead of skill or ability...

if so, the direct answers to questions probably dont matter... just have a ton of knowledge and computing logic ability in whatever languages or mechanics are required and confidently shift the topic to that... kinda like high school essay questions. i never knew the answer they were looking for cuz i didnt read or even have the book... but apparently that didnt matter cuz i explained the basics of the topic at a higher level than whatever the specific and typically dumb answer actually was.

maybe just try that.