1. 13
  1. 8

    This makes my brain hurt, what conclusions are we supposed to take away even if we did believe in their methodology?

    Oh R knocked C# out of the top five, not good news for .NET devs. HTML checks in at #16 we’d probably better not use it until it breaks the top ten. PHP is better than Javascript it says so right here.

    1. 4

      This is the single biggest issue with rankings of that kind, be it TIOBE or otherwise. It simply doesn’t convey much information other than what everyone felt already. Okay, the last part is a bit of a stretch, but still. This only gets worse when people pull out this as an “objective” argument in a language debate.

      Time to get my “The TIOBE index is not a credible source” shirt printed.

    2. 5

      SQL at #24, just ahead of Haskell? Oh-kay.

      1. 2

        As a DBA, I can assure you that very few developers actually engage with SQL, preferring to hit the DB through an ORM or other SQL wrapper instead. On the one hand, it’s frustrating, because there are still plenty of those “looping over ORM calls generates 10k queries where one would do” situations. On the other, it’s often fairly easy to tune something if they’re willing to embed raw SQL in their code (usually raw SQL that I write, mind you).

      2. 3

        Interesting, in this ranking, Rust is on place 26 for “spectrum”, and place 18 for “trending”.

        1. 5

          More strangely: Rust isn’t marked as being usable for embedded programming.

        2. 4

          Is it just me or is this list painful? C, Java, Python, C++, R, C# , PHP, Javascript, Ruby, Go

          1. 10

            Painful to whom?

            1. 8

              This might come as a surprise, but a lot of good and middling software is written in bad languages.

              1. 6

                More generally, almost all software is written in bad languages.

                (Also, almost all languages are bad. If you’re selective about who you ask, all languages could probably be bad.)

              2. 8

                Not surprising in the least. I don’t see why this is painful. Established, well supported, widely deployed and compatible languages lead an industry popularity ranking. Knowing that commercial applications and project have lifespans of many years or decades, this is not really a shocking observation.

                Going out on a limb to use the flavour of the month language in a large, important project that involves multiple developers is a risk:

                • No indication if the language will be alive and supported in N years.
                • Hard to find experienced or skilled developers.
                • Harder to find libraries.
                • Less mature tooling.
                • Lack of commercial support options for tooling, runtime, libraries, …

                I wouldn’t necessarily pick this for a hobby project or something personal. But I also wouldn’t go to my boss and tell him “hey man, I’ve been playing with this new Rust thing in my spare time, it’s really cool, perhaps we should rewrite our horrible 15-year old C++ codebase in Rust instead.”.

                1. 1

                  Strange to think that you don’t believe that C++ will be alive and supported in N years.

                  1. 4

                    Can you blame them for being an optimist? :)

                    1. 1

                      I think you misread. I’m pretty sure C++ will be alive and supported in N years for reasonable numbers of N (I’d say at least up to 15 years). If you don’t, I’m willing to take a long bet on it, for a concrete definition of “supported”.

                  2. 3

                    Interesting that R does so well. It seems to have really broken out of a niche (or that niche has grown enormously).

                    1. 3

                      R is tought and used at many universities, often outside the CS curriculum. Many forget how this is a good ground for popularity.

                    2. 3

                      Compromise is painful ¯\_(ツ)_/¯

                      1. 1

                        I certainly got no problem with any of them. They are successful for good reasons.

                      2. 2

                        I look at rankings like this as a sort of year over year relative indicator, nothing more. If I don’t know Swift I’m not going to be persuaded to learn it simply because it’s heating up on this list, though I expect many devs will do just that.

                        At worst it’s marketing for languages.

                        In the end it’s just a neat list to look at for a minute. Don’t think it’s supposed to be taken too seriously (is it?).