- Linus Torvalds criticized a Google contributor on the Linux kernel mailing list for his suggestions about filesystems.
- The debate centered around the use of inodes as unique identifiers for metadata on a filesystem.
- Torvalds’s strong language and tone in his response to the contributor were reminiscent of his past flippant attacks, for which he previously apologized.
Archive link: https://archive.ph/LP24s
You copied that function without understanding why it does what it does, and as a result your code IS GARBAGE.
AGAIN.
[…]
Debate continued for some time, in a cooler tone, with Torvalds offering suggestions on what he felt would be a better approach to the issues Rostedt hoped to address.
Harsh tone (in only two instances?), but he still invested in offering suggestions 🤷
I expected more behind the verb “flaming”.
Well, “YOUR CODE IS GARBAGE” seems like flaming.
You, send the same message without this and it achieves exactly the same thing without the “taking it all out on someone” part 🤷
Disagree. The hostility expresses frustration and indicates recognition of a long term pattern, and emphasizes the gravity of Torvalds’s observation.
Don’t get me wrong, I’m not pro-abuse – while the hostility does contribute to the content, that doesn’t mean it will have the desired effect. Some people get a fire lit under them when they experience that kind of feedback; others get defensive or heated or hurt, and those people are NOT motivated effectively – and it can cause lasting harm.
The key of it is: group B will not learn and grow from this approach; they’ll learn and grow from a more cool headed approach. Group A, however, will grow from either approach. Therefore, the cool headed approach is preferable.
I expected more behind the verb “flaming”.
He stopped doing that a few years ago.
Meh - I’m pretty sure Torvalds is just saying in public what thousands of other people were thinking quietly.
It sure is unpleasant to have your mistakes pointed out in public… but it’s a hell of a lot better than not even knowing you made a mistake at all which is usually what happens.
It would be better if Torvalds told the guy he’s an idiot in a private email but I’m not going to get worked up over that. Honestly I have a bigger problem with The Register making a headline out of it. The kernel mailing list is relatively private… this article is going to be attached to this poor engineer for the rest of his career. They should have omitted his name at least.
The kernel mailing list is relatively private
I thought it was public, just relatively unknown. It’s been a long time since I’ve looked it up
I’m assuming that’s why they added “relatively”
On the bright side, it looks (based on what I can tell) that things calmed down and the guy he was talking to took it fairly well.
If Linus “flamed” me, I’m putting it on my resume.
Breaking news: Linus Torvalds sends another email
Seriously though, I love to see the latest Linus Torvalds shouting match.
Boom, Rostedt.
I’m pretty surprised it was this guy; he’s pretty senior and has a ton of experience.
Linus could have shown a bit of leniency this time.
From what I’ve seen: Linus didn’t ever typically rip into first time contributors, it was usually always the people that he either expected better of or people that just kept submitting crap after crap.
Still sucks to be on the receiving end of it though.
Linus does that only to experienced guys - not beginners. People who should know better.
I would have liked a link to the LKLM thread. Usually they are quite informative
I’d better read something on the inode uniqueness problem than this. I’ve already got in trouble with GNU tar because of duplicated file identifiers in overlayed FS.
Angry Linus is the best.
