Live data from Hacker News

Why I Left Rust

jntrnr.com

221–230 of 880 posts

Re: Why I Left Rust

#221
"Rust acted as a cruel, heartless entity that did not care about JeanHeyd and treated him as disposable. "

"But it was just a downgrade. I shake my head at people that say things like this. Clearly, they are not used to treating people - let alone experts in the field - with respect. "

It was actually just a downgrade - and using terms like 'cruel' lacks proportionality.

It's a bit petty, especially for these kinds of public grievances over very personal, pedantic kinds of things.

Obviously what Rust did was 'not good' - these things happen all the time - and they need perspective and context.

All this huff and puff from people who take themselves a bit too seriously, maybe to the point of arrogance - we're professionals not artists, and that means 'making sausage as best we can'. It's all sausage, nothing is perfect, toes get stepped on - roll with it - that is the sign of maturity and confidence. If there is a systemic issue take that up.

Paradoxically it's these kinds of public slap fights over that make me wary of being engaged with a community, it's too much Kardashian.

Re: Why I Left Rust

#222
post #57
post #27

Earlier quoted context omitted.

Similar things might happen in other communities but Rust drama seems much more visible, and this isn't because of HN bias - the foundation having weird guidelines, controversy around governance, prominent people in the community badmouthing other languages in bad faith, now this. It's quite ugly.

I also think part of it is rust’s unique and complex governance model. Python, Go, C++, Ruby and Nodejs all had (mostly) a single leader who was in charge of the project for the first few years and who set the tone for the decisions which came after. If there was drama like this in Python, Guido would step in and that would have been that. Rust has no leader - and as far as I can tell, it never had a leader. Rust is…

It illustrates the weakness of this model of governance, with no clear leader. Politics can tear a project apart, and even ill-intentioned actors that do not want the project to do well can infiltrate and cause divisions. When a leader/founder is the BDFL, they usually want the project to keep flourishing according to their original (or evolving) vision.

Re: Why I Left Rust

#223
post #77

Wait. Let me get this straight. Someone is an expert in this field. They're asked to speak at RustConf after a leadership vote. They've also written an article about reflection in Rust - a purely technical thing that is already pretty widely disliked conceptually. (EDIT: the talk was about this, but it's also compile time reflection and came with the usual disclaimer that it was not representative of any of the Rust…

> The Rust project needs to stamp this out before it begins to fester. https://www.theregister.com/2021/11/23/rust_moderation_team_... There seems be something rotten at Rust and I've no idea what but anyone who gets a close look at Rust leadership seems to sprint away. It doesn't seem to be a matter of stamping something out it'll be an entire overhaul.

From Rust team discussion at Reddit, it's not that anyone sprints away, the answer is that no one has the time to balance volunteering for Rust, working on code, reviewing, doing technical review, and then on top of that moderation and doing diplomatic stuff.

Essentially no-one wants to do diplomatic stuff, so it falls to only person that likes doing that stuff (which isn't a good thing). Which can leads to episodes like this.

Re: Why I Left Rust

#224
post #77

Wait. Let me get this straight. Someone is an expert in this field. They're asked to speak at RustConf after a leadership vote. They've also written an article about reflection in Rust - a purely technical thing that is already pretty widely disliked conceptually. (EDIT: the talk was about this, but it's also compile time reflection and came with the usual disclaimer that it was not representative of any of the Rust…

Come on, Rust committee. Let's grow up here, shall we

Rust may, or may not, be the next best thing. However, the zealotry expressed by some rust pundits, is very off putting to me. One thing I have learned, is that zealotry is where moral compromise enters, it is where "for their own good" and "the ends justify the means" starts, followed by concepts such as "we need to force people to understand".

Some rust pundits seem to be such zealots, injecting rust commentary into everything. Almost religious.

Thus, I am not surprised by this. To speak ill of the holy relic, to utter dissent, drives zealots to mad excess!

Re: Why I Left Rust

#226

Earlier quoted context omitted.

If leadership acts so carelessly with regard to matters concerning a human being in such a shady and disrespecting manner, why would they act less careless with regard to technical matters? Will they just enforce their biased opinion, ignoring valid and generally accepted technical arguments? Rust leadership sounds a bit like the elite leadership of a communist state. I'd prefer an open dictatorship over that.

Notable in this backlash that this happened Friday evening/Saturday morning. So many of whom you’re characterising as the “elite leadership of a communist state” might not have seen this yet. They need to coordinate, build consensus and speak with one voice. I expect that might take a couple of days so they might have a response by Wednesday or so. Also, my understanding is that the consensus among the committee was…

No, I am not presenting this as a careful, considered decision. I was just responding why this incident could possibly also reflect on technical matters (not saying that it actually does). I have no clue what went on exactly. But clearly leadership somehow failed in this incident. Looking forward to hear more about how exactly this has happened.

Re: Why I Left Rust

#227
post #89

I guess one real benefit of rust is that it will encourage C++ to find way to be a bit more safer. cpp2/cppfront is one such initiative, and there are a few other programming practices that encourage safety. But other than that, rust is just a "modern ADA", nothing more. It cannot easily get interfaced with existing UI API either. Having verifiable code is a niche thing, it is very important, but most programmers don…

C is a terrible language for beginners. It's full of footguns and restricted in different ways because the compiler had to once fit in a kilobyte of RAM or whatever. C programmers who defend how easy it is to mess up programs written in the language often go with "well if you knew how to program C well you wouldn't run into this problem, so get good" from what I can tell.

C++ does better in some ways, and much worse in others (i.e. the incredibly verbose language, the Turing complete templates, the documented standards that take years to be implemented by any real compiler). Both languages are excellent for their problem space, but terrible for beginners.

Rust isn't verifiable code, the language is far from formally verified. Just one level down into the standard library you'll find tons of unsafe{} blocks and other performance related trickery that invalidates the idea of pure verification. It does verify more things, like ownership, that other languages like C++ ask you to do yourself (i.e. remember to use move() because the compiler won't tell you if you don't) but that's just a drop in the bucket of program verification. It also produces better programs, in my opinion; the logs of programs written in other native languages quickly end up full of failed assert()a and memory corruption related crashes still happen to me on a weekly basis. Rust programs have their flaws (for example, being completely unable to deal with memory exhaustion from a language perspective) but I see a lot fewer Rust crashes and bugged our programs than any other language in that space.

I also reject the notion that Rust is hard to read than C. It's more verbose, for sure, but that same complexity is still present in all of those "simple" C programs. You can just use mut for every variable and reassign it, you don't need to do all those functional operations or shadowing that you see in many Rust code bases. You can't ignore Result or Option types like you can in C, but in C fork() can still fail regardless of whether you're forced to deal with it or not, integers are still cast regardless of explicit casting, and all of the hidden assumptions about structs, pointers, and lifetimes are still present, regardless of whether you write them down or not. C lacks the ability to express complexity you're supposed to know, but that makes it more difficult to read correctly, not easier. The biggest issue with Rust readability is the alternate syntax inside macros, but there are plenty of preprocessor statements that do very much the same thing. Microsoft had developed some very extensive frameworks that are completely strung together by preprocessor macros exactly because the language was too limited to do what they wanted (but they couldn't afford to invent a better language, yet).

I don't think Python is C like at all. It's a scripting language without static typing so it can't really get too complex in the first place. Syntax wise, I'd group is closer to PowerShell than to C. Where it can, it does insert complexity (lambda functions, list/dict comprehensions, object oriented design) but it's relatively simple because you don't deal with memory allocation, pointers, or concurrency yourself.

You may be interested in Zig. It seems to be to C what Rust is to C++. The language isn't completely finished yet, but it's getting there. It features many modern language features in a language that is very C-like, with a potential solution for the allocation problem built into the language standard.

Re: Why I Left Rust

#228

I haven't seen anyone bring this up, so I will. I read through the original blog post from JeanHeyd[1] and in no way do they mention anything related to being a person of color, why then does JT associate it with that? I have no idea about the inner workings of the rust leadership team and who they are even, but from the timeline described and from the original post, there's nothing the could be related to that. Jean…

"White" is rather emphatically not like other ethnic or racial labels. Throughout the history of the US, who counts as "white" has been redefined many times to suit the political goals of the people at the time. In the era of the founding fathers, the Irish, Spaniards, Italians, all considered not white. Then during various times, often driving by waves of immigration of some other ethnicity people were freaking out…

...Calling a PoC ignorant, lecturing them and imploring them to read about racism... whew...

Them having been on the other side of the saddle, I'd bet they're familiar. How about simply listening to what they have to say without instruction, seeing their perspective and taking it in?

Re: Why I Left Rust

#230
post #88

Earlier quoted context omitted.

Much less weight should be given to someone "feeling uncomfortable" Rust the language is pretty good but the community around it has become really off-putting. The thick veneer of empathy and compassion quickly devolves into "idiot compassion" and emotional blackmail. Even the linked post contains the following: "I left because when I felt JeanHeyd's pain and disappointment at being mistreated and betrayed, my heart…

I thought "rust culture" came out of this sort of super SJW/woke culture of mozilla? Nothing against that sort of mindset in theory, but it sometimes leads to a certain irrational hypersensitive intolerant mob behavior and authoritarian/anti democratic tendencies. There was also the recent easily preventable drama [1] around the trademark guidelines some people left rust over. I don't really find any of this drama co…

[flagged]
Post reply on HN