Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Applying the Linus Torvalds “Good Taste” Coding Requirement

medium.com

Post reply on HN
Print view View original

221–230 of 302 posts

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#221

Post by stirner »

dghf wrote: post #70
stirner wrote: post #43

The author doesn't mention that the grid being initalized is square, and simply describes it as grid[rows][cols]. However, all his solutions are only correct for a square grid. One solution for arbitrary dimensions could be for (i = 0; i

You're updating the corners twice. Not really a problem with a simple assignment, but as no_protocol pointed out elsewhere [0], it could be if you ever wanted to change that to, say, an increment. Replace your second for statement with for (i = 1; i and you're golden. [0] https://news.ycombinator.com/item?id=12794235

You're right, my version is unoptimized because I was trying to keep the code neat as in the article. I think that optimizing the code for hypothetical future reuse is a bit premature. You could make a performance argument, but this optimization is going to prevent just 2 memory accesses, which aren't going to matter much compared to all the rest. If I was chasing performance I would start by memset'ing the top and bottom edges since they're contiguous in memory.

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#223

Post by jeremiep »

colanderman wrote: post #118
akkartik wrote: post #13

I dunno, the first example seems unsatisfying. The original has that ugly condition, the "good" version seems overly clever. And for all the talk of taste and aesthetics, both versions ignore an elephant in the room: defensively dealing with `entry` being absent from the list. Not that I've never abused addresses like this. But having written this multiple times, I currently prefer something like this: remove_list_en…

I don't think anyone's mentioned that there are two functional benefits to Linus's version: testability and performance. The author touched on testability. Dropping the "if" case eliminates one code path that may contain bugs. But also, the revised code may perform a lot better if you're searching a lot of short lists. In that case, the target item has a large, though not likely, chance of being first in the list. Th…

I'd say branch mispredictions and cache misses are the two biggest factors to performance. Almost everything else is noise in comparison. I keep saying if your architecture doesn't account for these, you've wasted 90% of the CPU guaranteed.

I do the exact same techniques in my game engine (sadly proprietary) - even in JavaScript its dead easy to run circles around both Three.js and Babylon.js in terms of performance to the point I'm simply amazed at how fast JavaScript VMs can be if you really push them.

Eliminating branches, cache misses and allocations in the main loop yielded by far the biggest performance boost. (For the curious, we abused ArrayBuffers to guarantee contiguous memory allocations in JavaScript)

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#224

Post by akkartik »

zAy0LfpBZLC8mAC wrote: post #107
akkartik wrote: post #13

I dunno, the first example seems unsatisfying. The original has that ugly condition, the "good" version seems overly clever. And for all the talk of taste and aesthetics, both versions ignore an elephant in the room: defensively dealing with `entry` being absent from the list. Not that I've never abused addresses like this. But having written this multiple times, I currently prefer something like this: remove_list_en…

Really, if you think the good version is overly clever, you probably have a very fundamental problem with your abstract reasoning. A very human one, as history is full of people having that problem--and ultimately, the good version taking over anyway. How about this: subtract(n){ if(n==0){ return m; } r=m; for(x=0;x People for a long time had strong objections to a number "0", because it doesn't make sense to have a…

".. if you think the good version is overly clever, you probably have a very fundamental problem with your abstract reasoning.."

Just to show that you're talking out your ass, here's a code snippet where I did the address of address thing in precisely the same "remove from linked list" situation. In a VM and programming language and operating system that I wrote from scratch[1].

https://github.com/akkartik/mu/blob/a90faae990/edit/003-shor... [2]

Moreover, I lived with it for months before I decided it wasn't worth the cleverness[3]. In this case. So I have some plausible claim to having found the "simplicity on the other side of complexity" in this case[4].

As a general rule I'd recommend not reaching conclusions about people from things they write in a comment on the internet. You're looking at them through a tiny peephole into their belief system. When in doubt, ask questions. "Excuse me sir, do you have much experience with C?", "Are you saying pointers to pointers are always bad?", and so on.

(Clever analogy there about 0. Totally agree with it. Utterly non-applicable in this instance.)

[1] https://github.com/akkartik/mu

[2] Hard to read, I admit. There's a colorized version at http://akkartik.github.io/mu/html/edit/003-shortcuts.mu.html, but I only host the most recent version there.

[3] https://github.com/akkartik/mu/commit/ea5e7fd4cb from Apr 2016. The previous version was from Sep 2015.

[4] https://en.wikipedia.org/wiki/Shuhari

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#225

Post by AceJohnny2 »

palunon wrote: post #117
AceJohnny2 wrote: post #8

These are great examples, and they hint to, but do not mention, the big counterpoint: development time. In his own examples, the author admitted that though the code was ugly, it worked. He then spent extra time reworking the existing code to make it, well, prettier. "If I had more time, I would have written a shorter letter." -- Voltaire The problem is that, in many (most?) professional settings, the developer is un…

"If I had more time, I would have written a shorter letter." -- Voltaire You're quoting Pascal, not Voltaire ;)

Zut!

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#226

Post by AceJohnny2 »

crdoconnor wrote: post #25
AceJohnny2 wrote: post #8

These are great examples, and they hint to, but do not mention, the big counterpoint: development time. In his own examples, the author admitted that though the code was ugly, it worked. He then spent extra time reworking the existing code to make it, well, prettier. "If I had more time, I would have written a shorter letter." -- Voltaire The problem is that, in many (most?) professional settings, the developer is un…

>Some projects, I daresay mostly open-source projects, can afford to be detached from the pressure of deadlines that corporations require That's partly why open source has staying power and so many corporate projects get junked.

Absolutely.

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#227

Post by akkartik »

crazy2be wrote: post #36
akkartik wrote: post #13

I dunno, the first example seems unsatisfying. The original has that ugly condition, the "good" version seems overly clever. And for all the talk of taste and aesthetics, both versions ignore an elephant in the room: defensively dealing with `entry` being absent from the list. Not that I've never abused addresses like this. But having written this multiple times, I currently prefer something like this: remove_list_en…

Another issue both versions have is that they will not tolerate head being NULL. Don't be afraid of double pointers, although they may seem overly clever at first glance, they can be really helpful for situations like this. Linus's code is considerably shorter, and not particularly difficult to debug or to understand, assuming you understand double pointers. Another case where double pointers can be really useful is…

I have nothing against double pointers: https://news.ycombinator.com/item?id=12798604 (please don't consider the flamey bits there to be directed at you).

That last bit, though, I wouldn't recommend it as a slogan :) How about something like "adapt to the situation for life"?

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#229

Post by Dylan16807 »

twwl wrote: post #138
sidlls wrote: post #111

Earlier quoted context omitted.

At some point that breaks down though doesn't it. I mean from one perspective a conditional in high level code doesn't describe "what is actually happening, either." That's especially true if an optimizing compiler or interpreter mangled it up. From another perspective your argument can be applied to any function call in library code. In either case I don't think your position is that strong.

> I mean from one perspective a conditional in high level code doesn't describe "what is actually happening, either." Sure, and from another perspective even the most efficient code does nothing to stop the heat death of the universe and is therefore functionally equivalent to doing anything else or nothing at all. However, that's just splitting hairs. It's not really an argument anyway, but rather a preference.

The question is more whether it's a reasonable preference to think syntax X is helpful but syntax Y is hiding things. Thinking deeper might show that the issue is actually familiarity, and the two should be rated the same in terms of abstraction.

Re: Applying the Linus Torvalds “Good Taste” Coding Requirement

#230

Post by Noseshine »

majewsky wrote: post #159
Noseshine wrote: post #95

Earlier quoted context omitted.

You make claim after claim, and I think they do need substantiating evidence. After all, I find it highly unlikely that if he were even 10% as bad as you make it sound he would be the head of one of the most successful projects - especially since everybody working with him does so voluntarily. There would have been breaks long ago. The fact that the Linux kernel held together under the original author loudly speaks a…

I won't argue with your main point, but a fair share of kernel devs are probably not working there voluntarily, since they write kernel code for their employer.

But this is backwards! You don't become a kernel developer by being assigned to the job by your employer. Instead, employers hire people who are kernel developers. Source: I myself contributed a teensy tiny bit to the network stack loooong ago (2.4 kernel), and I worked with real kernel developers when I worked at one of the major Linux companies and for them with large (mostly US) software companies.
Post reply on HN