I can't speak for him, but based on another comment further down ("I'll be over here in not-node"), it read like a dig at Node itself by proxy / extension.
> Yes, let's take the opportunity to condescendingly insult their mental competency.
Indeed. This type of behavior is almost never justified, but particularly less so when the object of scorn is clearly an exceedingly talented programmer. One glance at her GH profile reveals several well-used open source projects, including a neural network library and a bayesian classifier.
It really does drive home the point that if this can happen to someone of her caliber, then it really can happen to anyone. In fact, it probably does - except with less public attention.
Everyone writes crap code sometimes (viz. github.com/007 aka craptown USA). Just because you are "exceedingly talented" or have "well-used" projects doesn't mean everything you write is rainbows and unicorns.
If your project is useful to you, then screw everyone else - why care what they think?
https://github.com/harthur/kittydar is useless but brilliant. I'll be bookmarking it for the next time I need feature detection in JS, and because it mentions HOG descriptors.
https://github.com/harthur/replace is useful but not-brilliant. I'll bookmark the 'sed' or 'awk' manpage instead.
Thanks so much for pointing that out (and thanks Heather for writing it :P)! I would have missed it otherwise, and I have always wanted to play with neural nets as well as bayesian filters.. this seems very friendly for me to get into, so it made my week(end) if not more :)
The basic algorithms you need can be found via google.
In my (extremely limited) experience the much more difficult part is using them to successfully solve real world problems.
IMO the problem comes when you have people who's entire ego is based on being smart. In the past they probably had their egos massaged by getting the highest test scores in class etc and "winning" in that sense. Since the world does not work like that anymore they feel that they have to "win" at github instead.
Example: http://article.gmane.org/gmane.comp.version-control.git/5791...
goes ahead and provides one link everybody here has read by now
does see or mention context
But even if it's the case, he is rude with the guys who work with him, his "captains" that he knows since a long time.
He do not bash a random project on Github totally unrelated to him just like that.
You can't compare.
Linus is DFL (not so benevolent sometimes) of Linux, and can lash out when one of his trusted lieutenants is acting like an idiot. What these guys did is call out a stranger on the street and said "YOU SUCK!"
He's usually rude to code presented to him, but he doesn't go out of his way to diss code. It's a big difference.
He goes after people who enter discussions without bothering to understand the details or the history. If you want to code your own kernel in Java, that's your prerogative. But if you come to lkml and start preaching C++ advantages, you will get flaked.
If you don't understand the difference of context, well, look again!
(It's not necessarily good, but it is categorically different behavior.)
a. Linus uses swear words. Steve Klabnik used none.
b. Linus provides substantive criticism. Even if you need flame-retardant underwear to hear it.
c. Linus never ever ridicules outsiders.
So what do the two have in common, again?
I loved the mauro thread because you're seeing Linus talking to an insider of an influential project and tell him that the relative outsider with no status is right. By doing it publicly he reinforces the culture of the kernel: we don't break user code. We are a meritocracy, and if you're wrong I will not be shy about letting you know.
He could do it with less profanity. But I heartily support his doing it so emphatically. Nobody else on lkml will brush something off as a userland bug for a year.
There are many ways to reinforce meritocracy, or reinforce a culture without telling people to shut the fuck up in all caps.
That's actually not true. His "security monkey" references towards other projects being an example.
"I find it incredibly hard not to be judgmental.
I'm not sure what part of my personality makes this happen, but even when I try to curb tearing down other people, I end up doing it anyway. I'm not sure if it's just me, but I think part of the problem is that these kinds of things are rewarded."
http://blog.steveklabnik.com/posts/2010-09-24-trolling-is-a-...
His most recent post is http://blog.steveklabnik.com/posts/2013-01-23-node
> Twitter makes it so hard not to accidentally be an asshole.
No, I'm pretty sure being an asshole makes it so hard not to accidentally be an asshole. Plenty of people, including Heather, weren't assholes on Twitter.
"I'm sorry If I made someone feel bad"
Not
"I am sorry. I feel terrible that I made someone else feel terrible."
Corey's apology did feel nicer, but I don't think that Steve's was a bad apology.
I know that a lot of people do that. And I find it incredibily useful, sometimes finding a snippet of code saves me time from jumping extra hoops, just because (for example) driver misbehaves or documentation is incorrect/incomplete.
I hope this incident doesn't stop anyone from sharing their code and findings, no matter how good or "professional" they are.
> @harthvader @zeeg nothing's _wrong_ with it, but I don't want to build my app on top of others' code who are at this level of understanding
This message [1] in particular is absolutely disgusting. Someone hacked together a script that met their needs, and threw it up onto GitHub in case anyone else found it useful? Yes, let's take the opportunity to condescendingly insult their mental competency.
What does he even mean by saying that he doesn't "want to build my app" on top of harthur's code? Nobody asked him to; steveklabnik himself made the original post, taking the opportunity to show off this smug sense of superiority.
[1] https://twitter.com/harthvader/status/293829635823792128