Stories
Slash Boxes
Comments

SoylentNews is people

posted by cmn32480 on Saturday August 22 2015, @04:03PM   Printer-friendly
from the java-is-in-the-lead-because-of-the-caffine dept.

Think of it as a map of the rapidly changing world of computer software.

On Wednesday, GitHub published a graph tracking the popularity of various programming languages on its eponymous Internet service, a tool that lets anyone store, edit, and collaborate on software code. In recent years, GitHub.com has become the primary means of housing open source software—code that's freely available to the world at large; an increasing number of businesses are using the service for private code, as well. A look at how the languages that predominate on GitHub have changed over time is a look at how the software game is evolving.

In particular, the graph reveals just how much open source has grown in recent years. It shows that even technologies that grew up in the years before the recent open source boom are thriving in this new world order—that open source has spread well beyond the tools and the companies typically associated with the movement. Providing a quicker, cheaper, and more comprehensive way of building software, open source is now mainstream. And the mainstream is now open source.

Hmm, Perl has been declining...


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 4, Funny) by Subsentient on Saturday August 22 2015, @04:21PM

    by Subsentient (1111) on Saturday August 22 2015, @04:21PM (#226293) Homepage Journal

    C# and Java up, Javascript up, and C and python, my two favorite languages, declining.

    Kill it with fire.

    --
    "It is no measure of health to be well adjusted to a profoundly sick society." -Jiddu Krishnamurti
    Starting Score:    1  point
    Moderation   +2  
       Funny=2, Total=2
    Extra 'Funny' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   4  
  • (Score: 3, Informative) by N3Roaster on Saturday August 22 2015, @04:37PM

    by N3Roaster (3860) <roaster@wilsonscoffee.com> on Saturday August 22 2015, @04:37PM (#226299) Homepage Journal

    I wonder how much of C is declining is really just GitHub getting better at deciding what's C and what's not. I have a project hosted there that used to be detected as about 90% C when it was really mostly C++. Now it's showing that as only about 16% C (still an over-estimation) and it's estimating JavaScript far too low.

  • (Score: 1, Interesting) by Anonymous Coward on Saturday August 22 2015, @04:39PM

    by Anonymous Coward on Saturday August 22 2015, @04:39PM (#226301)

    C may eventually be replaced by Golang, which was created by some of the same people. Go wasn't on this list because not enough people are familiar with it yet. It probably will be in a few years (Docker, for example, is written in Go).

    • (Score: 2) by Subsentient on Saturday August 22 2015, @05:45PM

      by Subsentient (1111) on Saturday August 22 2015, @05:45PM (#226315) Homepage Journal

      I hope not. In many places, Go lacks features that C has. I do recall pointer arithmetic being among them...

      --
      "It is no measure of health to be well adjusted to a profoundly sick society." -Jiddu Krishnamurti
      • (Score: 2, Insightful) by Post-Nihilist on Saturday August 22 2015, @10:02PM

        by Post-Nihilist (5672) on Saturday August 22 2015, @10:02PM (#226410)

        Pointers arithmetic is something best left to the compiler, you only need to be able to get and set a pointers; even for system programming and GO provides that. When you think about using pointer arithmetic you should be thinking about slice and struct. Go ahead try to find a case where structs and arrays are insufficient...

        --
        Be like us, be different, be a nihilist!!!
        • (Score: 1) by massa on Sunday August 23 2015, @06:18PM

          by massa (5547) on Sunday August 23 2015, @06:18PM (#226698)

          Pointer arithmetic is good to write memory allocators.
          Really efficient string manipulations depend on really well-written string allocators (not present today on many "most-used languages", including C++, go, perl, ruby, python, etc. -- and don't get me started on Java...).
          But otherwise, yeah, C++-like iterators and for-each are more than good enough to substitute C-like pointer arithmetic.

          • (Score: 1) by Post-Nihilist on Monday August 24 2015, @12:18AM

            by Post-Nihilist (5672) on Monday August 24 2015, @12:18AM (#226768)

            Can you point me to an example where the following construct : address_of(array_u8[arithmetic]) would be insufficient? I am curious

            --
            Be like us, be different, be a nihilist!!!
            • (Score: 1) by massa on Monday August 24 2015, @02:25AM

              by massa (5547) on Monday August 24 2015, @02:25AM (#226786)

              Not insufficent, less efficient. The compiler has to have a clear view that you are doing

              for( i = 10; i 20; ++i ) prod *= a[i]

              and that it's the same as doing

              for( p = a+10; p a+20; ++p ) prod *= *p

              (which would translate to a single "repeat 10 times multiply addressed by register with autoincrement" instruction inside the loop -- those things in the core of the loops make the big difference in performance)

              (notice that in most cases, yeah, the compiler can do that. But the pointer arith is there to deal with the other :D cases)

              • (Score: 1) by massa on Monday August 24 2015, @09:22AM

                by massa (5547) on Monday August 24 2015, @09:22AM (#226944)

                (smart people know that there is a < that was omitted above, on each code example)
                :D

  • (Score: 2, Disagree) by ikanreed on Saturday August 22 2015, @04:54PM

    by ikanreed (3164) Subscriber Badge on Saturday August 22 2015, @04:54PM (#226305) Journal

    All of those but javascript are languages that have something positive to say about them.

    • (Score: 1, Informative) by Anonymous Coward on Saturday August 22 2015, @06:10PM

      by Anonymous Coward on Saturday August 22 2015, @06:10PM (#226329)

      What is wrong with javascript? It does exactly what it was built for, and well. Sure the plain old javascript has a lot more to go, but libraries built for it are making it magic for frontend development. jQuery and now Angular JS have freed legions of frustrated web developers from rendering web content on front end using bulky backend languages. As more and more people realize this, the portion of web code that used to be written in Java/PHP/Python that rendered the front end will be replaced.

      • (Score: 2) by ikanreed on Saturday August 22 2015, @06:31PM

        by ikanreed (3164) Subscriber Badge on Saturday August 22 2015, @06:31PM (#226335) Journal

        Yeah, you're right. Pointless digs at javascript don't hold up like they did in 2008.

      • (Score: 2) by Nerdfest on Saturday August 22 2015, @07:12PM

        by Nerdfest (80) on Saturday August 22 2015, @07:12PM (#226349)

        I remember reading about one of the creators of JavaScript saying that it was not intended for large scripts. They considered something like 1000 lines of code more than it was intended for. I'm not trolling here, I actually do a lot of JavaScript and don't *mind* it if it's written properly, that being with lots of unit tests (QUnit), and use of Require.js for dependency and namespace management. TypeScript is probably worth a look as well, or will be when a bit better tool support is in place.

        I really can't imaging people doing anything more than tiny projects without that. Dynamically typed languages require unit testing but I'm the only one I know that actually does it. Without something like Require.js *everything* pollutes the global namespace leading quickly to unmaintainable code.

        With those caveats it's not a bad language if you avoid some of the stupid inconsistencies.

      • (Score: 2) by DNied on Sunday August 23 2015, @05:34AM

        by DNied (3409) on Sunday August 23 2015, @05:34AM (#226559)

        What is wrong with javascript? It does exactly what it was built for

        Making the web a bloated mess and a security nightmare?

      • (Score: 1) by massa on Sunday August 23 2015, @07:40PM

        by massa (5547) on Sunday August 23 2015, @07:40PM (#226711)

        What is wrong with javascript?

        1. The Name. JavaScript is NOT Java
        2. console.log(typeof null); // object
        3. console.log(NaN === NaN); // false
        4. Global variables
        5. console.log(.2 + .4); // 0.6000000000000001
        6. Scoping is wacky
        7. Equality operators == and === are kludgy and inconsistent
        8. Lots of dark corners and unexpected behaviour
        9. Lots of reserved words, some of them not used in the language
        10. prototype

        And the list goes on and on and on... There wouldn't be CoffeeScript, Dart, TypeScript &c if JavaScript was not a deeply flawed language (that gained its traction from the popularity of its vehicle -- the web)

  • (Score: 2) by meisterister on Saturday August 22 2015, @07:00PM

    by meisterister (949) on Saturday August 22 2015, @07:00PM (#226344) Journal

    Welcome to the Glorious Garbage Collected, Interpreted/Jitted Object-Oriented Language Cult... joooiiiin.... usss.

    --
    (May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.