Today Linus Torvalds told a Google engineer that his code is garbage | Hacker News Reader

Today Linus Torvalds told a Google engineer that his code is garbage

twitter.com

13 pointspyman13 comments

I strongly dislike the spin of the headline – „Linus told a Google engineer that…” as opposed to „Linus told a developer that…”.

What is the headline implying? That Google engineers are gods? That Google engineers don’t make mistakes?

Also, reading through the code, Linus does have a point. I’m with Linus on this one.

It's the first sentence of the tweet. The engineer Linus emailed is also the Chair of the UNIX-Class Platform Specification. Anyway, the point is to discuss why Linus thinks his code is garbage, not the tweet's title.

And where is the news? "Google engineer" is not a status, it is a role. They can be fired tomorrow and they will be "developer". I know one Google engineer and they are not gods, by any means

Linus has a point – it was garbage.

I’m happy that Linus takes his job seriously and catches such things even under tremendous stress. We are lucky to have him; he once again saved our collective asses from a lot of pain down the road.

The tweet was prejudiced and manipulative. The bottom part of the email was cut off, not revealing additional context. The focus was put on „Google engineer“, which is totally irrelevant. People here in the chat repeat that the developer is „Chair of the UNIX-Class Platform Specification“. So what? Why is this relevant at all? Apparently, even „Chairs of this and that“ still have a lot to learn.

Another point: I have seen many such „helper“ functions when dealing with LLM-generated code. Was this particular code LLM-generated? I don’t know for sure, but it certainly looks that way. So the headline could easily have been „Linus rejects LLM-generated garbage.“

Once again, we are lucky to have Linus, whatever his communication style may be.

Interesting that this is worthy of comment. The inference I draw from the use of the words "Google engineer" is that Google and/or the tweeter now have a culture of deferring to authority, the job title, not measuring merit.

Doesn't bode well for Google.

He's not wrong, though, certainly not in the example he gave.

Yeah, but what's really going on here? Palmer is the Chair of the UNIX Class Platform Specification and has apparently been working on a RISC-V spec since 2019.

I love Linus' rants. I hope he keeps them coming.

He's right btw, it was garbage.

This is the kinder, more empathetic and approachable Linus, right ?

;)

Yes, and I'm not being ironic. The earlier Linus would have called the coder garbage, where here he's only insulting the code. I cannot emphasize how big of an improvement that is.

Do you think the way he reviews PRs sets a bad example for engineers who've just graduated and look up to him? I mean, there are different ways to reject a PR, even if the code is garbage.

Or maybe he was upset because he expected a lot more from someone who works at Google and has been working on RISC-V since 2019?

I don't know Linus, so I'm not sure what to make of this.

Yeah, it's probably not a great example, but you have to put it into context. Also, whatever problems there may be in Linus's communication style, are just about general professional communication and not specific to code review.

I read the message again to understand the actual issues being discussed. There are two: (1) the patch came too late in the merge window, and (2) the patch adds an unnecessary and obfuscating helper function. I don't have an opinion on (1), but I think Linus is completely right about (2). Calling it garbage is pretty harsh but not really wrong, so I wouldn't even say it's particularly over-the-top.

> Or maybe he was upset because he expected a lot more from someone who works at Google and has been working on RISC-V since 2019?

I don't know Linus, but I doubt "works at Google" carries much weight, and frankly that's an odd thing to focus on. If I had to guess, he probably read the patch, decided it was garbage, and wrote that in a message, not caring about where it came from. That's what I would have done (except I probably wouldn't have explicitly called it garbage, because I'm not Linus and that's not my style).

Reply on news.ycombinator.com