1. 3

    I’d really like to see some evidence that elixir is the most popular tool for “high-load” dev.

    1. 6

      Maybe not Elixir, but Erlang is pretty popular: https://twitter.com/guieevc/status/1002494428748140544

      “90% of internet traffic goes through Erlang-controlled nodes” and Cisco ships 2M Erlang devices per year.

      Elixir is in a certain sense just some developer friendliness on top of the Erlang VM. Still, most popular for application development? Probably not, but I’d say definitely mature enough to handle whatever you want to do with it.

    1. 3

      Great write up! Always nice to hear where one has been, and where they plan to go by inference.

      I got into all this hell we call programming via an iMac G3. Lime green. It was running OS 9 when I got it. I was a real Mac nerd back then. The worst part about it was considering myself a “Mac gamer” and how there wasn’t anything wrong with that. Ha. Anyway, I played RTCW and you could drop the in-game console down and make your gamer tag different colors etc etc via commands. I also played around with HTML and basic FTP servers.

      Then came my wilderness years spent in art school. There was a first year course that actually taught basic HTML and general web literary, which some students really disliked. Man, what a blast that all was. I got serious about a career in the arts before coming back down to Earth later and drifting back to basic IT work. Blah, blah … learned to program in earnest and here I am.

      1. 3

        Nothing to be ashamed of with regard to being a Mac gamer in the 90s and early 00s.

        Some of my favorite games were played on my G3 Beige tower with OS9.

        Also check out Richard Moss’ The Secret History of Mac Gaming

        1. 1

          Thanks for sharing this!

      1. 4

        Rust as a name was widely derided when it first came out. People pointed out that the name ‘Rust’ is a name that is invariably associated with the oxidation of iron, something that has negative connotations everywhere with decay, lack of maintenance. ‘But it’s named after a fungus’ really doesn’t help with that image either.

        Swift was also a funny one. When it came out, at least, it was anything but! It was slower than CPython in many simple numerical tasks, like simple for loops, and its type inference engine had some nasty bugs that lead to massive exponential slowdowns in performance.

        It annoys me when people say things like ‘programmers are notoriously awful at naming things’. It’s true that they are notorious for this, but it’s not actually true that programmers are awful at naming things at all. It’s nonsense dreamed up by marketing people that think that taking words that end in ‘er’ and making them end in ‘r’ instead is clever naming.

        1. 1

          Fair point. What I had in mind was things like variable and class names, API endpoint names…that I’ve personally encountered.

        1. 16

          I find the whole concept of ‘we may modify this at any time without your knowledge or consent’ clauses offensive. Nobody that feels the need to include such a clause is operating in good faith.

          I’ve heard people claim that this is them ‘covering their arse’ in case they accidentally modify the policy/terms or violate them in an unimportant way, but that’s exactly he opposite of what I want! How does that even make sense? Someone so careless that they would violate their own terms of service/privacy policy should be able to be held responsible.

          Those clauses should absolutely be illegal if they aren’t already. I also think that the courts should be much harsher on those that include unenforceable and illegal clauses inside such agreements. You shouldn’t just be able to throw the strongest possible legal terms into your clause and then have them watered down by the courts at a later date. Including anything unenforceable should render the entire terms unenforceable for the company. That way, people would be required to have terms of service and privacy policies that actually describe what your rights are. No ‘binding arbitration’ clauses that you just have to know are unenforceable. No ‘you give us the right to your firstborn child’-style clauses.

          1. 5

            TOS and privacy policies exist to control the behavior of end users, not the company providing the service.

            It doesn’t even make sense to talk about a company “violating” their own TOS because the TOS doesn’t apply to the company, but to the user. The company can always change the TOS and Privacy Policy to suit their whims.

            1. 1

              Do you know if Privacy Policies and/or TOS are legally binding? I’ve heard they are not.

              If not, what could a digital product provide that would lock both the provider and customer into some set of agreeable terms? This sounds like a contract to me, but I’ve never seen this done in any sort of digital product.

          1. 9

            GitHub is a wonderful product. It’s incredibly well designed despite the complicated nature of the workflows it supports. The fact that the app is still usable and perfomant with JS disabled is a significant achievement. I do worry that they will go the way of JIRA and come out with some fashionable but low-usability react-based modal-everywhere redesign. That would be unfortunate.

            1. 12

              the app is still usable with JS disabled

              Indeed! Now if only it would be usable with JS enabled; that would sure be nice.

              (The way the JS in their comment forms intercept common readline-based and emacs-based shortcuts and replace them with useless markdown formatting functions is so annoying I had to blacklist their JS.)