@TeachMeCode that's precisely how he got this position in the first place

Whatever makes you happy!

@jschmold

> I much prefer what I'm used to

Good for you! What is it?

> real wonky language with too many hidden intricacies

Care to name a few?

> I wanna read code and "reason about it" without having to lookup a hundred other functions…

That's odd, my experience is quite opposite.

There's a write-up from a yesterday's keynote talk at fby.by by Michael Snoyman explaining why Haskell actually _is_ awesome - https://snoyman.com/blog/2017/...

Been there years ago. As a pretty dumb person, I can assure you it is comprehensible. As an added bonus, it actually lets you do more with less resources

Need a hug?

@irene well the business logic is high-level, and c++, among many other, actually lacks sum types (no, std::optional is months old and probably won't make 90% of existing projects; no, std::variant might or might not land C++ 2020) which forces you emulate them with trees of ifs and cases carefully handling errors with exceptions when you actually need goto, or some AbstractVisitorFactory nonsense you'll forget next day (not sure which is worse).

Consider this instead:
https://fsharpforfunandprofit.com/p...
Designing with types: Making illegal states unrepresentable

@irene low-level - probably, high-level - probably not.

@irene and I'm barely scratching the surface. Just consider this - functional programming is *very* different, so any assumptions you may have from previous experience are likely incorrect. Mine, for one, were (mostly because being self-taught).

@irene still enough for 99.99% applications

@Mayhem93 why, not all of them are shit. Just these mainstream ones I've worked with for last fucking decade and have so fed up with them and computer abolitionist assholes defending the right to assign a human to the job computers were fucking invented for over and over again. /rant

@irene it's not *just* runtime check, the unused code paths are eliminated by a compiler by trivial graph traversal. Which can't be trivial in imperative languages, 'cause all the code being impure by default.

@irene what do you mean by cleverness exactly?

@Mayhem93 how having a brain is an elite thing?

Mercurial got it properly anyway ;)

@vkubre ya, that feeling of being in control is warm and fuzzy, except you need decades to be actually in control with c++

@irene by being lazy and not doing things for needed to be done

Only doing it manually, with synchronization primitives.

A wild computer abolitionist appears!

But what?

@Hevlastka what's both funny and frustrating is when you want Git to do something non-trivial you're on your own

Me too. I've switched to Mercurial after very frustrating half of a year with Git and that was so relieving even back then in 2011. When you're reading manuals because you want to, not because you have to. When a tool talks human, not bullshit gibberish can't be red without glossary.

Consider programming languages with sum types

> segmentation fault
Enjoying technologies from 70s yet?

@filthyranter the unnecessary complexity of Git is still there, along with its' design flaws

@filthyranter no it doesn't. You are getting used to.

Doing software development on Linux since 2008, am rebooting to Windows for games only.

Mercurial + MutableHistory - Git done right

Consider openSUSE Tumbleweed: we actually test it logs in after upgrade (and many more). And get this: these tests are automated, so no humans need to be paid for QA

@lostinmyworld well, *anything* is simpler than Git, so that won't take so much resources as one might think.