Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Linux

The Linux Foundation Releases Annual Linux Development Report 101

Posted by samzenpus from the naming-names dept.
darthcamaro writes "The Linux Foundation's Who Writes Linux report (sign up required) is now out and after 22 yrs leading Linux, Linux creator Linus Torvalds has fallen out of the list of top 100 developers in terms of code contributions. He currently ranks 101st for number of patches generated from the Linux 3.3 to the Linux 3.10 kernel releases." Read below for a few highlights from the report.
Nearly 10,000 developers from more than 1,000 companies have contributed to the Linux kernel since tracking began in 2005. Just since the last report, more than 1,100 developers from 225 companies have contributed to the kernel. In fact, more developers and companies are contributing to Linux than ever before with Linux kernel 3.10 seeing the most developer contributions ever.

Mobile and embedded companies are increasing their investments in Linux. Linaro, Samsung and Texas Instruments together increased their aggregate contributions from 4.4 percent during the previous version of the paper to 11 percent of all changes this year. Google’s contributions are also up significantly this year.

The Top 10 organizations sponsoring Linux kernel development since the last report include Red Hat, Intel, Texas Instruments, Linaro, SUSE, IBM, Samsung, Google, Vision Engraving Systems Consultants and Wolfson Microelectronics. After appearing on the list for the first time in 2012, Microsoft notably dropped off the list entirely this year. A complete list of the top 30 organizations sponsoring this work is included in the paper.

The rate of Linux development is unmatched. The average number of changes accepted into the kernel per hour is 7.14, which translates to 171 changes every day and more than 1,200 per week.
This discussion has been archived. No new comments can be posted.

The Linux Foundation Releases Annual Linux Development Report More

The Linux Foundation Releases Annual Linux Development Report

Comments Filter:

  • Yes but... (Score:5, Interesting)

    by Meditato ( 1613545 ) on Monday September 16, 2013 @12:29PM (#44864881)

    does that matter? He still maintains the repo, still performs the merges, still does the quality control, still determines the direction of future updates. As Slashdot is fond of saying, the quality of a developer isn't just determined by the number of lines or commits he contributes.

    I don't know why that bit about Torvalds is even necessary, unless someone is trying to take a swipe at him. Again.

    • Re:Yes but... (Score:4, Insightful)

      by BitZtream ( 692029 ) on Monday September 16, 2013 @12:34PM (#44864935)

      Or you could just take note that Linus is becoming more of a manager than a raw code producer.

      Of course, this is completely normal and part of the process a developer goes through in their lives as they progress through their stages.

      Its just something to note, no need to get all offended, especially since we're not even talking about you, Fanboy. Linus will eventually stop writing code for Linux all together for any number of reasons, including the inevitable death.

      Pull your panties out of your crack and move on. No one is insulting your God.

      • Re:Yes but... (Score:5, Insightful)

        by Anonymous Coward on Monday September 16, 2013 @12:44PM (#44865027)

        I'm with you.

        To me, the shocking thing is that it took him 22 years to fall out of the top 100. I would have assumed he'd have fallen out AT LEAST 10 years ago - for the very reasons you cite.

      • Its just something to note, no need to get all offended, especially since we're not even talking about you, Fanboy. Linus will eventually stop writing code for Linux all together for any number of reasons, including the inevitable death.

        Pull your panties out of your crack and move on. No one is insulting your God.

        Calling someone a fanboy with all the abusive tone of Linus Torvalds. How ironic.

        Just FYI, I'm not at all attached to Linus or the manner in which he conducts himself towards others...but the continual complaints and passive aggressive swipes can be even more tiring and petty than the man himself. Linus is just not all that big a deal. In what world is the statement "so-and-so has fallen out of the top 100 contributors, ALL THE WAY to 101st!" news? Why are we even talking about him, again? That's my questio

        • "Calling someone a fanboy with all the abusive tone of Linus Torvalds. How ironic."

          Yeah. I don't think you get the difference. When Linus gets "abusive" as you call it, the person on the receiving end almost always deserves it. That is fundamentally different than what is going on with this guy.

          " the continual complaints and passive aggressive swipes can be even more tiring and petty than the man himself. "

          Linus is actually what we would commonly refer to in the intellectual world the exact opposite of p

          • The Internet worked fine before Linux existed, BSD has more to do with it than Linux.

            My phone runs Mach/BSD.

            My Depth sounder/GPS runs vxWorks.

            My router at work runs Cisco IOS.

            My router at home runs FreeBSD (pfSense).

            Every DNS server I connect to is FreeBSD or Solaris, though I suspect Solaris won't be around much longer at the ISP side.

            My ISP either uses cisco devices, juniper devices (FreeBSD based OS), or FreeBSD based devices such as BigIP load balancers.

            My DVR runs Window Media Center and whatever the X

            • "The Internet worked fine before Linux existed, BSD has more to do with it than Linux."

              Great. Now conjugate to the proper tense and you'll see where you went horribly wrong.

              "My phone runs Mach/BSD."

              Hey! That's awesome! (though doubtful). Of course, you are conveniently leaving out that it is useless without another one on the other end, as I already pointed out.

              "My ISP either uses cisco devices"

              A) By your own admission, no access to your ISP for you (or didn't you know Cisco uses Linux) B) Now that you g

          • Try to use the internet without using his code. Oh wait ... you can't.

            Well damn it, did netcraft finally confirm *BSD is dead? Did it take IIS and Apple-server with it too?

            Call someone to complain that you can no longer use the internet then ... that should work! Damn, you or the person you were going to call has an Android phone!

            Yeah, because we didn't have either custom phone OSes or WinCE / WinMobile before. Or if in your "non-linux" world Apple wouldn't dominate the phone market... if Apple was the only phone OS maker I'm sure they would license o/sX to anyone who could prove decent hardware compatibility just to grab as much money as possible; app store purchases alone would be astronomical orders of magnitude larger that w

            • "Well damn it, did netcraft finally confirm *BSD is dead? Did it take IIS and Apple-server with it too?"

              Allow me to explain how the internet works! Your packets have to get from your machine to the BSD based server or IIS based server and back again. If you could run a cable directly to those servers, you'd be golden, but you can't now can you. That means your data gets handled by Linux even if it is served from a BSD machine to your Windows box.

              Ok. Now that we have straightened out that little cogniti

              • No, yours is just off the wall knee-jerk bullshit.

                Let's break this down into little words you can understand:
                Your claim is any device currently running Linux / Embedded Linux right now wouldn't exist in any way shape or form if Linus hadn't made the kernel.
                My claim is any device currently running Linux / Embedded Linux would have an analogous device running another form of OS / Embedded OS if Linus never existed.

                Lets examine a little deeper:

                Ok. Now that we have straightened out that little cognitive distortion you were clearly experiencing, lets look back to everything else you wrote, modify it to the present tense, and see if you can still say any of it. Go ahead. I'll wait.

                Yes, because clearly all of the Linux based devices just vanish

                • "Your claim is any device currently running Linux / Embedded Linux right now wouldn't exist in any way shape or form if Linus hadn't made the kernel."

                  I never made any such claim, and you are a bona fide moron. I didn't bother to read the rest of your idiocy. When you pass an elementary school English course you will learn the difference between can't and couldn't.

                  • Well you definitely need to go to the zoo for the Gorilla translation. I'm sorry, I can't afford to smash my head into a wall until I'm as stupid as you.

                    Please do have a good day sir, the stupid ones are always the happiest; you have that to look forward to.

            • > if Apple was the only phone OS maker I'm sure they would license o/sX to anyone who could prove decent hardware compatibility

              A few years ago, Apple had the only credible smartphone OS. They didn't license their smartphone OS when they were the only one. Instead they gave the majority of market share to a company who DID license their OS.

              Is that a mistake they would only make once? For several years they had the only GUI OS for desktop computers. Rather than license it, they left every other manufactu

      • Re: (Score:2)

        by Kjella ( 173770 )

        Or you could just take note that Linus is becoming more of a manager than a raw code producer.

        Of 1100+ developers he's in 101st, that's what still in the top 10%? Granted many of those won't be full time but many of them will too, some of them developers that do nothing but crank out code. The real story here is that the day has only 24 hours and with more full time developers joining up Linus is pushed down. Not to mention more developers means more code is pushed upstream to the subsystem maintainers and eventually to Linus so there's more to review as well which is also a form of digging into the

      • Re: (Score:2)

        by Lennie ( 16154 )

        Or Linus has more and more merges to do, so less time to code.

      • "Or you could just take note that Linus is becoming more of a manager than a raw code producer."

        The number of participants continues to grow and the commit rate continues to accelerate. It therefore doesn't follow that he is becoming more of a manager. He could increase his output and still "fall" to number 101.

        "Of course, this is completely normal and part of the process a developer goes through in their lives as they progress through their stages."

        You are confusing being a benevolent project dictator wi

        • The number of participants continues to grow and the commit rate continues to accelerate. It therefore doesn't follow that he is becoming more of a manager. He could increase his output and still "fall" to number 101.

          What? You're telling me that being the guy who does all the merges ISN'T managing? Or that from now until the end of time, there will be faster and faster committers? That makes no sense. Adding more committers doesn't make him slower or them magically faster.

          You are confusing being a benevolent project dictator with holding a job at a typical company.

          So I'm guessing you've never managed a software project in your life. Managing a project is managing a project, be it open source or ultra secret proprietary NSA work. You seem to have some silly fantasy about Linux being magically different than

          • "What? You're telling me that being the guy who does all the merges ISN'T managing?"

            There isn't a person on the planet who understands software that calls that managing.

            " Adding more committers doesn't make him slower or them magically faster."

            It makes perfect sense. More samples means you might still be on the same point of the curve, but the number of points above you on the curve will increase. You are just unable to understand how statistics work.

            "I can not offend you, but you or he can take offense. "

          • I can not offend you, but you or he can take offense. Thats your problem, not mine.

            That's just a way of dodging responsibility for your words and actions. You can offend somebody if you mean offense, and you meant offense. Internet pseudo-anonymity doesn't change the fact that you were acting like an asshole to someone whose motivations you have no idea about (FYI: far from fanboy). Get off your high horse.

            Despite my previous response, you still haven't told me how exactly I was being a fanboy. There is precedent for silly passive aggressive swipes against Linus Torvalds on Slashdot...rem

      • Or you could just take note that Linus is becoming more of a manager than a raw code producer.

        Of course, this is completely normal and part of the process a developer goes through in their lives as they progress through their stages.

        Its just something to note, no need to get all offended, especially since we're not even talking about you, Fanboy. Linus will eventually stop writing code for Linux all together for any number of reasons, including the inevitable death.

        Pull your panties out of your crack and move on. No one is insulting your God.

        Do you always blow a tampon when someone expresses their discontent with an author's presumed attitude? Did I effectively express that you are a bitch for being so insulting?

    • Its probably not a swipe. Just acknowledging that Linux is primarily developed via corporate/governmental subsidies and not the more romantic hobbyist developer contributing his/her personal time.

    • does that matter? He still maintains the repo, still performs the merges, still does the quality control, still determines the direction of future updates. As Slashdot is fond of saying, the quality of a developer isn't just determined by the number of lines or commits he contributes.

      Actually I think Linus has enough work with all that. He probably wouldn't even have time to do much coding.

  • Greatest (Score:5, Insightful)

    by Anonymous Coward on Monday September 16, 2013 @12:30PM (#44864889)

    There is no other person who has led as large and successful a software project for as long as Linus has with as much involvement as Linus has. I think that pretty much makes him the greatest software development manager of all time. It also means that those who criticize his management style need to pony up more than just their opinions.

    • Re: (Score:1, Interesting)

      by BasilBrush ( 643681 )

      The first contrary example that quickly springs to mind:

      Guido van Rossum started Python in 1989.
      Linus didn't start Linux till 1991.

      And Guido seems to do it without being an asshole.

      I'm sure there are others in the open source world, and many more, though perhaps not well known, doing long standing closed source projects.

      • Re: (Score:3)

        by mrvan ( 973822 )

        Python seems to have around 750kloc (approximately 50/50 c and python code), while the linux kernel has 17mloc. That is more than 20 times as much code.

        This does not mean Linus is doing a better job managing than Van Rossum, but it certainly says something about the respective complexity of the projects.

        • Lines of code is not a good measure. It's a very constructive commit if you replace 30 lines of code with 3 that do the same thing. Yet by a LOC measure that would be a retrograde step.

          Or look at it another way, suppose you had to projects that do the same thing. Say two C compilers that both meet the latest spec, and produce equally efficient code. But one had 10 times the LOC as the other. The better one would be the smaller one.

          17 MLOC for a kernel is ridiculous. Even Torvalds admits it's far too big. On

      • Re: (Score:1)

        by gl4ss ( 559668 )

        guido not an asshole? he put in python all the whitespace shit. he is a gigantic asshole.

        there's plenty of sw projects that have been going on for far longer for sure but not any that has had the same impact, even if there's even some that fill almost the same purpose.

        the less the software matters the easier it is to "not be an asshole" about it. besides, it's just management by perkele. he hasn't been a real asshole to anyone - if you want you can ask a politicians scriptwriter to translate the sentences i

      • "The first contrary example that quickly springs to mind:"

        That would be an excellent contrary example if it were not for the fact that it doesn't come close to qualifying as an example. First of all, you can't possibly be claiming that Python is as large a project, or as pervasive. Secondly, you are confusing the fact that most people have never heard of someone because they are very low profile with the fact that people will get jealous and call a famous person they have never met an asshole.

        "I'm sure the

        • First of all, you can't possibly be claiming that Python is as large a project, or as pervasive.

          Largeness of source is a flaw of the Linux kernel not a plus point. And pervasiveness does not relate to the claim made.

          Yes. I'm sure there are many projects that are bigger, and of course they all compile and and are usually run on ... oh shit! I was really pulling for you not to sound like an idiot, too.

          As you're a bigger asshole than Torvalds, your opinion is worth nothing to me.

          • "Largeness of source is a flaw of the Linux kernel not a plus point. "

            So you are saying you have no idea what you are talking about. I can accept that. The largeness of the source is due to the fact that it targets more than 30 architectures, and has thousands of features, each which can be compiled in or left out based on the needs of the target. Furthermore, one can choose from hundreds of tuning options.

            "As you're a bigger asshole than Torvalds, your opinion is worth nothing to me."

            Thank you! I canno

            • So you are saying you have no idea what you are talking about. I can accept that.

              On the contrary, even Torvalds says it's getting too big. So you're the one who doesn't know what he's talking about. But then you are an asshole.

              • "Largeness of source is a flaw of the Linux kernel "

                Is English your second language? Calling it a flaw isn't the same as Linus saying it would be easier to manage the codebase if it wasn't so big. As usual, you are nothing but a clueless wanker who cannot read and/or understand even the most basic language constructs and concepts.

                • as Linus saying it would be easier to manage the codebase if it wasn't so big.

                  Which wasn't what he said, asshole.

                  • I don't blame you. I'd be embarrassed and trying to deflect the attention elsewhere if I made the kind of statements that you made too. To save you time and face, don't worry about it. I accept your apology.

    • There is no other person who has led as large and successful a software project for as long as Linus has with as much involvement as Linus has.

      ...
      GNU > Linux

    • Re:Greatest (Score:4, Interesting)

      by elashish14 ( 1302231 ) <`profcalc4' `at' `gmail.com'> on Monday September 16, 2013 @08:30PM (#44868937)

      I think that pretty much makes him the greatest software development manager of all time.

      Automatically, when anyone makes these kinds of statements, I consider Stallman as well. Linux wouldn't have gotten very far without a compiler, and my understanding is that gcc was basically the only compiler around at the time (you know what I mean - icc, etc. don't count). And of course, you could also make a claim that it was all of Stallman's work with GNU that kept Linux and many other projects open instead of being overtaken by greedy interests and left to die in obscurity.

      While all of this makes for a great debate, it's of course always going to be very subjective and there's never any real answer. I'm sure that there are many others that could also go in here as well (Larry Wall, Ritchie, Kernighan, and so on).

    • Humm, in terms of Open Source, you're probably right but I think you'll find that most guys looking for career advancement wouldn't hang around just doing merges and maintaining control of the project ad-nausea. I think what you're seeing here is the fear that if Mr. Torvalds steps away from his project it would fail, the contrary should be true that if he does step away that there are others that can fulfill what he's doing now. To each his own but if I was doing the same repetitive thing over and over f

  • He should really be called GNU/Linus.

  • That's gotta count for something.

    • Re:But Linus still ranks 1st in profane tirades (Score:5, Interesting)

      by geek ( 5680 ) on Monday September 16, 2013 @01:33PM (#44865527)

      I wouldn't say he's first. He's just the most publicized. Theo from OpenBSD is pretty bad and a quick glance over the Debian mailing lists will blow your mind. Linus is bad but there are far worse offenders than him.

      • There's lot of flamage on Debian mailing lists, but next to no profanity.

        I delete messages quickly, but in 648 I currently have left, there are 2 fucks and 2 bullshits (not counting a false positive in a GPG signature).

  • Nobody from Ubuntu (Score:5, Interesting)

    by Drunkulus ( 920976 ) on Monday September 16, 2013 @01:23PM (#44865425)
    No mention of Canonical anywhere in the report. Why am I not surprised?

    • Re:Nobody from Ubuntu (Score:5, Insightful)

      by Gavagai80 ( 1275204 ) on Monday September 16, 2013 @02:22PM (#44865977) Homepage
      Ubuntu's supposed purpose is to add polish, so one shouldn't expect them to be doing kernel work.
      • Maybe an even more important point they contributed is a large community. Cannonical took many years to build that up by marketing, providing infrastructure (forums, launchpad) and hand-picking the various pieces software that makes an distribution of things that work well together. This has several advantages: First, more users means more testing, more bug-reports and more people that can help you in a forum. Secondly, a large community means they created a critical mass to pressure hardware vendors to rel

      • Re: (Score:2)

        by NotBorg ( 829820 )

        Ubuntu's supposed purpose is to add polish, so one shouldn't expect them to be doing kernel work.

        So what the fuck are they doing with Mir? Polishing a turd?

    • Re:Nobody from Oracle (Score:2, Insightful)

      by Anonymous Coward

      Canonical isn't going to have a kernel team like the competition; their focus is the desktop side. Oracle is the surprise failure. They've complained before that their expertise rivals Red Hat, but this report appears to disagree.

  • Wow - this seem such a niche market companies (with all due respect) for making top 10 contribution to linux kernel - interesting...

    • Semiconductor production is hardly a niche market. Sure, you might not hear about them much, but Sun, IBM, Lenovo, Intel, Samsung, Apple, Google, and all the others have to go somewhere for their hardware. Why re-invent the wheel at significant cost when TI and Motorola already have a production line for what you need?

      • Well, I have looked at Wolfson - they specialize in audio (yes including silicon). Still find it interesting that they contribute to linux kernel along IBM's and Googles of the world. And Vision Engraving? Is this the same company: http://www.visionengravers.com/ [visionengravers.com]?

    • Re: Vision Engraving and Wolfson ? (Score:5, Insightful)

      by tlhIngan ( 30335 ) <slashdot&worf,net> on Monday September 16, 2013 @02:31PM (#44866035)

      Wow - this seem such a niche market companies (with all due respect) for making top 10 contribution to linux kernel - interesting...

      No, it makes complete sense - these guys do drivers and such as a marketing exercise. When companies come to them, they want to answer "Yes, we have a driver for you, it's already in the kernel". OEMs are far more likely to choose a company that has drivers already in the kernel than not (and thus need to develop one). And being mainline In the kernel is a quality bar - it's one thing to have a driver to integrate yourself, it's another to have one that's in every kernel going forward.

      So a lot of these contributions are "scratching their own itch" where the itch is "sell more of our chips".

      • Re: (Score:1)

        by Anonymous Coward

        That's not such a bad thing is it? It means that Linux has gone main stream enough to create real demand.

      • OEMs ... are the ones who make drivers. OEMS are ORIGINAL Equipment manufactures.

        You're referring to integrators and VARs.

  • For dreaming up the project and executing it. Everyone else is a follower, no matter how valuable their contributions are.

  • This sounds like a metric about as useful as LOC (lines of code) - it favors all those devs who like to make multiple 2 line changes instead of checking in all related files together without breaking a build.

Slashdot Top Deals

"Hello again, Peabody here..." -- Mister Peabody

Close