rewriting apioforum #289

 
arcade src #2873

what if we did it in raw C cgi scripts

gollark src #2876

Implode immediately.

BlueManedHawk (edited ) src #2877

[snipped—see original here]

honestly these don't seem too unreasonable to me […]

Infinite punishment is always immoral. I've been meaning to write out a formal proof of this for a while, but i don't have enough space on my computer to install a TeX suite, so for now i'll just regurgitate what i always say: Because an infinite punishment must only be given for infinite crimes, and no jurisdiction is perfect, a jurisdiction must be prevented from ever giving an infinite punishment, because it is not okay to risk someone being given an infinite punishment for a finite crime.

As for a fair and just trial…well, this has been covered before, and i'm sure you already know why a fair and just trial is important, so i'm not sure why you consider it okay for the Rust community to allow punishments to be delivered without them.

A governing body punishing people for crimes committed outside of their own jurisdictions is not okay because it makes it impossible for a person to begin anew.

Better than object-orientation.

rust isn't object oriented lol

The language allows for object-orientation, and the standard library has a lot of it.

mb src #2878

new longest thread? /j

BlueManedHawk src #2879

RE: #2871

Yeah, i agree that these things suck. Hopefully, WG14 will be able to make some progress to change these things. Regardless, i think that C is still the best option.

BlueManedHawk src #2880

new longest thread? /j

What was the old one?

jmibo src #2881

opinions on rust:

i really like everything about rust other than memory safety (which i have mixed opinions about). it has a lot of features that should be in most languages (blocks can return values, mostly), enums are great, the error handling makes sense, it has what i find to be the right amount of object-oriented things, etc.

the memory safety is a bit of a mixed bag for me; while nobody likes getting a compiler error because you didn't figure out how to avoid using more than one reference to an object, nobody likes segfaults either, nor std::move and rvalue references and whatever happens in c++. but they could probably afford to make it a little less strict? i really don't know.

taswelll src #2883

you can use more than one reference to an object! as long as they are all shared (immutable) references

ubq323 (bureaucrat) src #2885

you can use things like cells and refcells and rc too (i forget what they are all called exactly)

amby src #2886

but they could probably afford to make it a little less strict?

iirc the safety checks will reject stuff that might be correct but it can't prove it, instead of accepting stuff that might be wrong but it can't prove it. because oracle problem

taswelll src #2887

yes, this is by design

please log in to reply to this thread