3 ms·

> your arguments against Rust come off as biased. Yes, I'm pretty aware of this, and I keep this bias voluntarily, because Rust is generally touted (hyped?) as

> your arguments against Rust come off as biased. Yes, I'm pretty aware of this, and I keep this bias voluntarily, because Rust is generally touted (hyped?) as a Silver Bullet, and I'm wary of Silver Bullets. So, I want to highlight that fact. > Sorry if I came across as rude. No hard feelings. As I said, I'm not under delusion of a perfect life. I just said it. > I just find it slightly puzzling how your stance on it is kind of going up and down. :) I have a habit of mirroring the tone of the person I'm talking with. Combine it with the fact that English is not my native language, it's possible that "tone adjustment" is far from perfect. > it's been quite a while since I've seen Rust zealotry on HN... Yes, there's no zealotry here, and it's great, but I'm exposed to a wider community than HN people. Unfortunately, HN represents very small percent of the programmers around. A great talk I like about this issue is Robert Martin's "What Killed Smalltalk Could Kill Ruby, Too" [0]. Let's say, I'm marred and bitter from tons of zealotry and flamewars over the years (yes, I'm not young folk). > While that likely does wonders for your mental health... [Snipped for brevity] First of all, thanks. I pray that you work in a place you love and never work again. I'm very aware that how being solo makes me different and biased, and I think I highlighted at a couple of places. On the other hand, I understand what teams and team-related activities are important for code quality. Let's say that I work as a team of two people. Current me knows everything about code, and my future me which inherits that code. So, I always code and comment for my future self, who doesn't know anything about the code. I always sharpen my axe, and try to improve myself as a "team of one". My personal state of the art of this practice is at [1]. I'll reflect what I have learnt from this project to the next one. The fact is while I do extensive C++ testing, I yet to pick up Go's testing abilities. That'll be the next step probably. > Tinkering is nice and all but I'd probably make your life living hell if I had to review your PRs. :D Give it a look at [1], and maybe [2], and let's have a chat again. I'm not perfect, for sure, but I think I'm no basement dweller when it comes to code and docs quality. :D > OK, that's valid but I'd still urge you to reassess your opinion of the Rust community. As a recovering grumpy, I'll do. I don't like to be bitter grampa (to be). I'm fine with the grampa part, but not with the bitter part. > No harm done IMO, and I don't think that I disagree strongly with anything you said. Same, no hard feelings here. I can say that you're good sport even, and I don't use this lightly. I enjoyed what you wrote and writing this comment. > I am mostly saying that you working what you love and working solo is keeping you quite disconnected from what are good team practices. Of course, but I'm trying to incorporate the good practices I can use as a team of two (as aforementioned). Also, I'm mingling with more Free Software teams than I might show. > Maybe you have a compiler in your brain as you have alluded to but I will trust a good test suite over a human brain every day. I'll trust my brain first, then evaluate that with the compiler, then evaluate the end product with a good test suite. Then feed what I learnt to myself, so I don't repeat the same mistakes as much as I can. My intuition gives me perspective, but I always operate on the assumption that I'm the worst programmer in the universe (no kidding). [0]: https://www.youtube.com/watch?v=YX3iRjKj7C0 https://www.youtube.com/watch?v=YX3iRjKj7C0 [1]: https://git.sr.ht/~bayindirh/nudge https://git.sr.ht/~bayindirh/nudge [2]: https://github.com/TRUBA-HPC/lssrv https://github.com/TRUBA-HPC/lssrv

pdimitar 3y ago
> I keep this bias voluntarily, because Rust is generally touted (hyped?) as a Silver Bullet, and I'm wary of Silver Bullets. Hrm. We're not progressing on that front so maybe it's time to stop. But I don't think Rust is "touted as a silver bullet". Rust is simply better than many other languages if you have criterias X, Y and Z -- and it just so happens that those X, Y and Z are actually things that fix problems that exist quite often out there. This is not "touting". Forgive the generalization but you do seem to belong to a group of people who get annoyed if others praise something and if they hear about it often. Granted, popularity is not a signal for quality (and we as senior techies should possess and practice this kind of critical thinking much more than many other people!) but this is also tech and I'd like to pretend that at least part of all decision-making processes are based on merits. ;) I'll give you a non-tech example. "Game of Thrones" was hugely popular and I was acting seemingly like you do now -- I kept hearing about it so much that I actually made it a goal to NEVER engage with it. Obviously that's a bit irrational, if not a bit damning for the person practicing such methods in their lives. But again, in tech it's not exactly the same. If you keep hearing about Rust, then critical thinking and objectivity demand that you ask yourself why and assess it with a clear head and without bias. You seem to claim to have (mostly) done so but reading into your seeming-rebeliousness against Rust, I have some doubts. Hence this call-out (and the thread between me and you itself). > Yes, there's no zealotry here, and it's great, but I'm exposed to a wider community than HN people. I understand, yet IMO you should make up your own mind and not get irritated just because something gains popularity. Sometimes there are good reasons for the popularity. Now in my day work I am pissed that the languages I work with don't have sum types. :\ Fixes so many bugs... > Current me knows everything about code, and my future me which inherits that code. You are surely aware that is not enough unless you actively go out of your way to check how other people do things, right? But I won't pursue this further, I believe we're in agreement and only the degree of stuff is under question from my side. > As a recovering grumpy, I'll do. I don't like to be bitter grampa (to be). I'm fine with the grampa part, but not with the bitter part. Well that's all that I asked for really. Thanks. > Of course, but I'm trying to incorporate the good practices I can use as a team of two (as aforementioned). And one of them should IMO be: "when you need something that you'd write in C/C++ and it is not highly specialized (meaning no good Rust library support) then try writing it in Rust". Several huge organizations came forward and said memory unsafety contributes to at least 70% of all zero-day vulnerabilities. Surely at this point being religiously fanatical about memory safety should be common sense, I'd think. > My personal state of the art of this practice is at [1] You didn't have to show me, appreciate it. I do like the well-commented code. But I'd definitely remark that a huge if/else should be broken down into two functions. Maybe the else clause even further -- talking about `if isInputFromPipe(logger)...` in `initFlags`. I'd like the code to tell me what is it doing, not describe it to me how it's doing (and I still don't know the "what" part while reading such code). But that kind of stems from Golang's imperative nature and not everyone wants to write code in mostly FP style. I get that. > Same, no hard feelings here. I can say that you're good sport even, and I don't use this lightly. I enjoyed what you wrote and writing this comment. It's important to disagree productively. :)

pdimitar 3y ago

> I keep this bias voluntarily, because Rust is generally touted (hyped?) as a Silver Bullet, and I'm wary of Silver Bullets. Hrm. We're not progressing on that front so maybe it's time to stop. But I don't think Rust is "touted as a silver bullet". Rust is simply better than many other languages if you have criterias X, Y and Z -- and it just so happens that those X, Y and Z are actually things that fix problems that exist quite often out there. This is not "touting". Forgive the generalization but you do seem to belong to a group of people who get annoyed if others praise something and if they hear about it often. Granted, popularity is not a signal for quality (and we as senior techies should possess and practice this kind of critical thinking much more than many other people!) but this is also tech and I'd like to pretend that at least part of all decision-making processes are based on merits. ;) I'll give you a non-tech example. "Game of Thrones" was hugely popular and I was acting seemingly like you do now -- I kept hearing about it so much that I actually made it a goal to NEVER engage with it. Obviously that's a bit irrational, if not a bit damning for the person practicing such methods in their lives. But again, in tech it's not exactly the same. If you keep hearing about Rust, then critical thinking and objectivity demand that you ask yourself why and assess it with a clear head and without bias. You seem to claim to have (mostly) done so but reading into your seeming-rebeliousness against Rust, I have some doubts. Hence this call-out (and the thread between me and you itself). > Yes, there's no zealotry here, and it's great, but I'm exposed to a wider community than HN people. I understand, yet IMO you should make up your own mind and not get irritated just because something gains popularity. Sometimes there are good reasons for the popularity. Now in my day work I am pissed that the languages I work with don't have sum types. :\ Fixes so many bugs... > Current me knows everything about code, and my future me which inherits that code. You are surely aware that is not enough unless you actively go out of your way to check how other people do things, right? But I won't pursue this further, I believe we're in agreement and only the degree of stuff is under question from my side. > As a recovering grumpy, I'll do. I don't like to be bitter grampa (to be). I'm fine with the grampa part, but not with the bitter part. Well that's all that I asked for really. Thanks. > Of course, but I'm trying to incorporate the good practices I can use as a team of two (as aforementioned). And one of them should IMO be: "when you need something that you'd write in C/C++ and it is not highly specialized (meaning no good Rust library support) then try writing it in Rust". Several huge organizations came forward and said memory unsafety contributes to at least 70% of all zero-day vulnerabilities. Surely at this point being religiously fanatical about memory safety should be common sense, I'd think. > My personal state of the art of this practice is at [1] You didn't have to show me, appreciate it. I do like the well-commented code. But I'd definitely remark that a huge if/else should be broken down into two functions. Maybe the else clause even further -- talking about `if isInputFromPipe(logger)...` in `initFlags`. I'd like the code to tell me what is it doing, not describe it to me how it's doing (and I still don't know the "what" part while reading such code). But that kind of stems from Golang's imperative nature and not everyone wants to write code in mostly FP style. I get that. > Same, no hard feelings here. I can say that you're good sport even, and I don't use this lightly. I enjoyed what you wrote and writing this comment. It's important to disagree productively. :)