From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964947Ab3GQWzr (ORCPT ); Wed, 17 Jul 2013 18:55:47 -0400 Received: from mail-pa0-f53.google.com ([209.85.220.53]:61996 "EHLO mail-pa0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964892Ab3GQWzh (ORCPT ); Wed, 17 Jul 2013 18:55:37 -0400 Date: Wed, 17 Jul 2013 15:55:39 -0700 From: Guenter Roeck To: Steven Rostedt Cc: Ricardo Ferreira , David Lang , ksummit-2013-discuss@lists.linuxfoundation.org, Greg Kroah-Hartman , Darren Hart , Ingo Molnar , Olivier Galibert , stable , Linux Kernel Mailing List , Linus Torvalds , Willy Tarreau Subject: Re: [Ksummit-2013-discuss] [ATTEND] How to act on LKML Message-ID: <20130717225539.GA2120@roeck-us.net> References: <20130716212704.GB9371@thunk.org> <20130716224357.GK4994@xanatos> <1374015299.6458.76.camel@gandalf.local.home> <20130716231217.GL4994@xanatos> <51E5E609.1030202@oracle.com> <1374022296.6458.118.camel@gandalf.local.home> <20130717055139.GC21611@1wt.eu> <1374063678.6458.124.camel@gandalf.local.home> <1374066215.6458.135.camel@gandalf.local.home> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1374066215.6458.135.camel@gandalf.local.home> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Jul 17, 2013 at 09:03:35AM -0400, Steven Rostedt wrote: > On Wed, 2013-07-17 at 13:30 +0100, Ricardo Ferreira wrote: > > Slashdot is just a cesspool of trolls, not a good comparison. > > Point taken. > > I posted this privately, and I think I'll repost it here. I need to > modify it a bit as it wasn't meant to be public. > > > When I started sending patches to LKML it was not the cursing I was > afraid of, it was the possibility of top notch developers pointing out > my flaws. Linux is intimidating not because it can be harsh, but because > its the big league. You are posting code not only to the world but also > to some of the best programmers on the planet, and frankly, that's > really scary. And I think that's the real reason people who are shy tend > not to want to participate. They use the harshness of LKML as an excuse, > but I think it's really that they may be insecure about their own work > and how it will compare with the best of the best. > > Both my wife and I have done karate for decades. My wife has even won a > national tournament. She can do demos without a problem, but when she > has to get up in front of other top black belts, she's a nervous wreck. > She's her biggest critic, but she tends to know that when performing in > front of people as good as she is, or better, they can see her flaws as > much as she can. That is intimidating. > > The point I'm making is that we need to find out what is preventing good > developers from joining the Linux community. Is it really the harshness > of the project, or is it because we expect you to have the best code, > and you will not be accepted if you are not that good. And I do not want > people joining that are not good programmers. > Preventing good developers from joining - I don't know. Maybe there just are not that many. I have heard lots of reasons for not paricipating in open source development. The "official" stated reason is often around "not exposing our IP", where "IP" is sometimes declared to be each line of code. Another is "we don't want to help our competitors". Personally I believe that being afraid is only part of the picture. Good developers should ultimately know that their code is good, and not be afraid to show it (or find a mentor to encourage them). However, I have to say that that much of the code I have seen in my life is _not_ good, or crap as is referred to by many in the Linux community. To some degree includes my own code - if I encounter code I have written ten years ago, I often think "did I really write this crap ?". I think _that_ is a key reason for people not participating - they are afraid that their code might be exposed as crap. A corrolary of that might be that some companies don't want their customers to see how bad the code is they are shipping to them. > The answer is not to bash Linus into being a nice guy (which seems to be > what Sarah's trying to do), but we can get mentors or even "scouts" to > look for people of talent and help them get into the community. What > those people need is not a nicer LKML that will let mediocre developers > in, but someone that recognizes their talent and encourages them to > join, by reinforcing to them how good of a developer they are. I've > helped people this way. Talented programmers that were unsure of > themselves, and they have done extremely well in our community. > Excellent summary. I absolutely agree. Thanks, Guenter