Hacker News Reader

Comment by DodgyEggplant | Hacker News Reader

Linus manages combined efforts of hundreds of people for years, probably leading the most successful open source project ever. Without salaries, without command and rule authority, without perks, without shared office space, even without HR. He probably knows something about what it takes to manage people

View on HN

Or maybe he's just so talented that he can carry a huge interpersonal handicap.

Actually, this email is pretty polite and to-the-point. I don't know exactly what it is that he said that was offensive in private, but he's actually being pretty reasonable here.

The thing is, some people spend a lot of effort when writing to check that they don't accidentally leave "sharp edges" in their writing (and then they still fail to find them all) while others write the way they would talk and assume that their tone of voice is "obvious" when of course it isn't. I expect if you knew Linus personally (I do not) his most outrageous written comments would seem just fine.

There's a ton of research on how written communications (especially with relative anonymity, etc.) can lead to cascading flamewars and so forth without anyone setting out to troll or be outrageous. Linus lives in that world, so he probably offends a lot of people accidentally just as a matter of percentages.

> most successful open source project ever

Surely that would be BSD which, without Linux, would probably be Linux (and was Linux before Linux), and with Linux is still most of Linux and a good deal of Mac OS X and iOS, and without which Linux would be irrelevant.

>I don't know exactly what it is that he said that was offensive in private, but he's actually being pretty reasonable here.

I think what's being discussed is Linus' reaction to someone's patch on the lkml a few days ago. https://lkml.org/lkml/2013/7/13/132

Here is a preview:

  >What the F*CK, guys?



  >This piece-of-shit commit is marked for stable, but you clearly never

  >even test-compiled it, did you?

Wow, I cannot tell you how many times I've wanted to say something similar in a code review but instead go with "hey, this looks a little funny, any ideas what you could do to clean it up?" Part of me does think its rude, the other part is jealous he can say it and I can't :)

Haha, yeah. Honestly, being nice in code reviews, half the time the person just argues for half a dozen comments or never addresses the issue, even if it's completely obvious like two methods doing the same thing but with different names. Oh, well, if we have to be nice and not get things done or suffer some idiot code getting in, we have to be nice.

If you're wired anything like a normal human being, you'd observe the effect saying it actually had and you'd stop behaving that way. So think of your current situation as having skipped ahead without the additional damage of a hands-on learning process.

"Hey, this looks a little funny, any ideas what you could do to clean it up?" is ambiguous and doesn't help you.

"Hey! This is broken. It doesn't do X, Y, or Z. Please make sure to check code thoroughly (refer to $DOCUMENT) before committing." is direct, tells people what's wrong, and isn't hostile. People will be offended by it, but those people can be made to understand that they are being given useful direct advice.

The feeling of being offended by such a message diminishes the more it happens, the more you fix and prevent the behaviour, and when you see your role models make similar mistakes, which has a tendency to humanize them.

Agree. I *hate it when I receive ambiguous criticism, which just means that we'll need to dance around each other while I try to figure out exactly what "looks a little funny".

Getting specific details - instructions or guidance even - is far better to solving the problem efficiently.

Both are really not the response you want to give, but the one you gave probably would be less problematic and, if said coder was paying attention, might get the right response. Code reviews with folks who cannot take a hint were always the least fun experience. I was a TA during college, so at least I had some training. I often think seniors should have to do a formal code review with freshman just to get the vibe.

That style has merit. Whomever submitted that code, which obviously resulted in creating extra work to waste Linus' time, that person has many choices, such as leaving and never contributing again, or improving their code before next submit which might actually help the project. If they choose to tough it out improve their code in future submissions, then they will likely also have improved themselves in a lot of ways. Otherwise, they can just go away, and at least in this popular project, there is a chance they will be replaced by a much better talent.

The code may be bogus, but that's not actually what set Linus off. The target of his fury is not the code's author (Kees Cook, who is unlikely to be driven off by a Linus rant anyway), but trusted lieutenants who missed an obvious problem in a serious process failure. Gleixner has been around the better part of a decade, and hpa's contributions date back to the early 90s.

This is a "You know better than this!" rant, not an attack on some new guy.

That's a good way to ensure that people will test-compile patches in the future before submitting.

I don't believe that quote is what she had an issue with.

See http://marc.info/?l=linux-kernel&m=137390362508794&w=2

If that is what started it off, then I feel she more or less either misinterpreted what was being discussed in that thread, or else is deliberately picking a fight.

The reactions he got were great. Everyone apologized and will now make sure such mistakes don't happen again.

Having been on lkml before, I can tell you, nearly everyone does it. And, yes, that includes Sarah Sharp:

Date: Thu, 26 May 2011 00:04:50 -0700 From: Greg KH To: Sarah Sharp Cc: linux-usb@vger.kernel.org Subject: Re: [RFC 0/3] xhci: Remove useless debugging

On Wed, May 25, 2011 at 04:28:51PM -0700, Sarah Sharp wrote: The xHCI driver is what, two years old now? It's high time it put on its big girl pants and stop crapping out useless debugging information.

    You now owe me a wet-wipe to clean my laptop up from the coffee I just
    snorted out my nose all over it.

From: https://www.linux.com/news/special-feature/linux-developers/...

The wind was probably blowing South East instead of North West or something on the day Sarah decided everyone needed to be more "professional".

There's a ton of research on how written communications (especially with relative anonymity, etc.) can lead to cascading flamewars and so forth without anyone setting out to troll or be outrageous.

Genuinely interested in this research. Care to cite?

I'm not an academic so I can't reel off citations from memory (my wife can but I don't have her handy).

Here's a quick example thanks to google:

http://www.wired.com/science/discoveries/news/2006/02/70179

But this is a drop in the bucket.

Actually, I reckon he's being fairly unreasonable, he's just couching it in superficially-reasonable terms. He's arguing that "playing the victim card" is a way of enforcing your expectations upon others, but accusing someone of "playing the victim card" is itself an incredibly nasty way of imposing your own expectations on others.

It's essentially saying that not only should Sarah Sharp be willing to put up with all the hostility, but that the only reason she's complaining about it is to make a political point. It's a way of snidely arguing that she can't possibly actually care about the issue. That's far more dismissive of other people's reactions than anything Linus Torvalds is complaining about, and yet he's trying to claim that he has the inclusive moral high ground and Sarah Sharp is the one who's ignoring the fact that different people react differently.

Totally agreed, accusing someone of "playing the victim card" is a super lame oratorical trick, frequently used by people who don't realize that that's what they're doing. It's up there with saying, "now, I know it may not be politically correct to say this...", which is pretty much 100% of the time followed by some completely idiotic statement. The trick is that it sets up an environment where objecting to the idiotic statement is preemptively made to sound like some sort of hypersensitive political correctness.

Reply on news.ycombinator.com