>>107451380 (OP)
>The harder the language is to use without fucking up, the better the programmers that use it are on average.
Only if a lot of the fucking up is caught by the compiler, like Rust, Ada, Haskell, Coq, etc. These are the kind of languages that keep bad programmers out.

>>107451380 (OP)
It's less about how hard the language is and more about how much it abstracts from raw data/io.

>>107451380 (OP)
When the iphone first came out, Steve Jobs and others in Apple debated for a long time over if they should allow third party developers or not. Requiring the use of Objective-C was seen as a compromise because it was so difficult to use that it would prevent the worst shovelware and crap from being viable. After Jobs died, Swift was created to make developing for iOS easy. The current state of things is the result.

>>107451380 (OP)
>The harder the language is to use without fucking up, the better the programmers that use it are on average.
wrong: see javascript, c++, php, and other
and it is also wrong in the opposite direction: go is easy to use and has produced and/or influenced lots of excellent programmers

>incompetent programmers flocking to the language
literally never happened. rust has a comically high entry barrier, specially if you come from c/c++ and want to do things like moves

>>107452218
WRONG
third party apps were a fundamental part of the business model from day 0, by the time jobs released the ipad they were straight up telling media companies that they should develop an app for the iphone and ipad because people were willing to pay for apps but not for physical media (https://archive.md/FGSTk). it turned out to be a lie, but a lot of media companies fell for it because of FOMO

>>107452218
>Requiring the use of Objective-C was seen as a compromise because it was so difficult to use that it would prevent the worst shovelware and crap from being viable.
That's like saying Objective-C was bad on purpose. C family languages tend to be "difficult to use" because of bad design. They are technical problems that are really people problems, in this case, the low IQ of their designers.

>>107451380 (OP)
No, you are out of your mind.

>>107452414
>go is easy to use and has produced and/or influenced lots of excellent programmers

Name one example?

>>107451380 (OP)
>Rust tries to bubble wrap everything, but this led to incompetent programmers flocking to the language
It didn't? Rust is a very difficult language. You don't see pajeets use it

>>107452414
>javascript
>he thinks javascript is hard
>it literally runs no matter what shit code you throw at it
Opinion discarded

>>107451380 (OP)
>incompetent programmers flocking to the language
jeets are literaly filtered by the borrow checker.

the cloudflare proxy was well behaved, it's good behavior to exit with an error when given a broken config file, nginx does it to.
the real issue is that 1. a bad file was generated, 2. it was being deployed without being validated to be correct and 3. it wasn't automaticaly rollbacked by the monitoring system / orchestrator.

it's outside the scope of responsability of a proxy to monitor system health and mitigate broken config files.

>>107454175
>>107451380 (OP)
btw rust is one of the harder language to learn out there, c++ is much simpler to learn in comparison, i do think once you master it you are much more productive with rust, but jeets are filtered and never get there, the compiler just won't let them shit everywhere like they are used to and they get frustrated and then make spam hate thread about the language with "arguments" that don't make sense because they don't understand shit.

>>107452725
>>107454187

this, i've used rust professionaly for the last 5 years, all my coworkers have been straight white men.

about 6m ago i started a job as a lead dev because they wanted me to fix their shit, told me i could use rust if i wanted but their codebase was mostly go.
first day meeting, i meet the whole team, literaly 12 jeets, i look at the codebase, utter spaghetti, as much as i didn't like go i had never seen such a bad codebase.
i left the job on the same week.

i'm never taking a job that even mentions go again as it's always a jeetfest.
now i'm happily doing rust again.

>>107454205
go was literaly made by google and designed for retarded jeets lmao

>>107454205
>i'm never taking a job that even mentions go again as it's always a jeetfest.
Really? The Go teams I've worked with were all Chinese and Koreans, they were all excellent. I live in a country with few Indians though.

>>107454251
interesting, idk, all go dev i've ever met in my career have been browns.

i live in switzerland if that's any relevance.

>>107454251
>>107454261
though the jeets i mentioned here >>107454205
were in a remote job, the company was located in germany

>>107452414
>WRONG
Learn some history boy, especially if you're going to call others out. The debate happened and Jobs was persuaded to allow others on the platform, but they had to use Objective-C. Swift didn't come until years after Jobs left. Your post is why everyone hate zoomers and their "the world only has existed since I've existed" attitude.
Before you reply, which you will because your fragile ego requires you to do so, re-read what I posted several times and think about what the words mean. Then punch yourself in the face for being such a zoomer retard.

>>107451380 (OP)
It's why I prefer writing C++ and manually managing my resources over the braindead python programming I've been doing for the past year. It drives you to think hard about what your code is actually doing so you deeply understand what's going on at a lower level. And that makes you a better programmer.

>>107454400
This is true. Writing high performance code is about actually understanding what the data you're manipulating looks like in terms of layout and size, and how it's being accessed and manipulated, in order to maximize cache efficiency. By far the biggest bottleneck in computers is the actual business of moving data to and from the heart of the CPU and a good programmer reduces this as much as possible. If you abstract away the business of dealing with memory, how can you write good data-oriented code?

>>107451380 (OP)
>apparently it processes new configuration, and crashed at the unwrap because configuration with too many features was passed in.
That jeet language lmao, I can't read it without a jeet accent.
Explain what unwrap does to someone who never used rust and why would it crash?

>>107454521
rust doesn't abstract it away much though.
if you know how computers work you know what rust is doing for the most part.

and you can always manually do everything if you realy wish to, the std is not magic.

>>107456106
It means getValueOrTerminate(). If the result has no value because an error occured, close the program.

>>107451380 (OP)
Ok, what's your point? What's the takeaway? That everything should be built in assembly?

>>107456106
rust is one of the language with the least jeets actualy they get filtered by the borrow checker, it's one of the harder language to learn out here but if you master it it's also quite productive.
and yea, unwrap is not complicated, basicaly a function that can fail will return a Result, the result either wraps its return value, or an error.
to be able to get the value inside it, you have to deal with the possible error, the standard way to do it is to send it up with "?" and manage it with a match statment up the call stack.

unwrap basicaly just does "if it's Ok ie not an error, return the contained value, if it's an error, exit the program with an error".

no programmign language will prevent you from writting "if this function failed, exit the program with an error" and pm all rust devs know what unwrap does;

>>107456152
Hard to have any jeets when nobody uses your language.

>>107456125
>>107456152
>If the result has no value because an error occured, close the program.
Why would you do that?
I understand you should use an assert if something is wrong (a bug which only a programmer can to fix) to stop the program to prevent further damage, but why would you terminate the program for something that can be checked and handled at run time?
And how could "too many configurations" cause an error that cannot be handled at run time???

>>107456194
> Why would you do that?
there are many scenarios in which exiting a program with an error is the proper behavior.
ie when started with a broken config file, that's what nginx does for ex.
if you want another example, if you give a blender file to a cli tool that expects images the proper behavior would be to exit with an error saying "the file is invalid".

> but why would you terminate the program for something that can be checked and handled at run time

the point is that not everything can.
and it's up to the dev to decide.
you don't have to use unwrap, it's a choice.
typicaly in rust errors would be handled with "?".
which is syntax sugar for match result { Ok(data) => data,
Err(error) => return Err(error)
}

the error can then be handled somewhere up the call stack.
you can also just use match, or map, or stuff like unwrap_or_default().
anyway, nothing in rust forces you to exit with an error when a function fail, but that can be the desired behavior.

>>107456194
>>107456216

>And how could "too many configurations" cause an error that cannot be handled at run time???
so now it's a question of scope of responsability.

1. the broken config should have never been generated
2. it should have been validated before ever being deployed
3. it should have immediately been rollback by the monitoring system / orchestrator upon failure.

it isn't the proxy's job to manage system health, it's the monitoring software / orchestrator, imo it did what it was supposed to do given a broken config, nginx has the same behavior.

>>107456192
you like to think that but hundreds of popular codebases uses rust.
and it's slowly getting everywhere.
even in the linux kernel (though i don't think mixing codebase is ever a good idea).

>>107456225
So why didn't those "white" "heterosexual" "intelligent" rust programmers make sure a wrong config is never generated?
Where did it all go wrong, jeet?

>>107456241
imo it's more to do with with bad upper managment.
the people reponsible for writting the proxy are probably not the same one that those responsible for the rest of the system.
cloudflare is jeeted, even if i don't belives the ones writting the proxy were jeets, it's not realy their fault if other part of the company are jeeted.

also, the one that deployed the change, saw that everything broke and didn't revert, the dude was probably a jeet and had nothing to do with the proxy that refused the file.

>>107456252
There's no doubt that cloudflare uses rust because it is jeeted.

>>107456266
i've never found a single jeet capable of using rust, go on the other hand ...

>>107456323
Oh so you yourself don't use rust?

>>107451380 (OP)
If C programmers are so smart, why did they ever think that gets() was a good idea?

>>107456325
i'm not a jeet.
i'm the guy that made this post : >>107401283 →