I read that one, he literally described himself as mediocre programmer and is excited about gpt as a way for mediocre programmers to be competitive again. I’m sure he’s in for a really fun time when he has to find a bug in 12k lines of AI spaghetti he bolted together.
I presumed it to be a standin for just directly using Math.max, since there’s no nice way to show that in a valid code snippet
Not using thief is professional incompetence unless you’re doing something deeply cursed
Fair enough, though I contend that for a common-case application like a database-backed REST API where the architecture is basically standardized there is no meaningful time difference between writing crappy code in a clean architecture and writing a crappy pile of spaghetti.
I’ve been tasked with updating some code a senior programmer (15+ years experience, internally awarded, widely considered fantastic) who recently left the company wrote.
It’s supposed to be a REST service. None of the API endpoints obey restful principles, the controller layer houses all of the business logic, and repositories are all labeled as services–and that’s before we even get into the code itself. Genuinely astounding what passes for senior-level programming expertise.
That’s what’s really irks me be about JS–you can do just about whatever but you’re not supposed to.
It’s an imperative language, but best practices are to use it functionally.
You can omit semicolons, but best practices are to use them.
You can use sloppy equality, but best practices are to always use strict.
It’s admittedly quite good at what it was originally supposed to be: a voice chat service for playing games that’s easy to join, use, and share. The troubles began when they started trying to pivot to be a general-purpose public internet space provider, because the platform was never supposed to be that and they’ve done absolutely nothing to support it.
Eh, depends on the language and the context. I still use 80 for C, but I’ve found 120 to be a much more reasonable number for Java.