1. 3
    We want to make Nix better nix determinate.systems
  1. 35

    Flagged; this doesn’t seem to be anything other than promotion for Determinate Systems plus “we will totally give to the community guide, promise”.

    More technical submissions, less marketibg.

    1. 6

      Not a nix person, so take this with the appropriate grain of salt. Having organizations work on these problems seems cool, but without a specific announcement about a new tool or approach, this post seems a bit light for lobsters.

      Nonetheless, hope you come up with something good.

      1. 4

        I’ve recently taught myself Nix and started using NixOS. This isn’t my first attempt and I believe is my third and clearly my most dedicated attempt. I’ve long understood the benefits Nix has to offer, but in the past I was unable to spend the innovation costs in my own workflow. I’ve worked with all the major Linux distributions both professionally and personally, and have experience using the configuration management tools built for them (e.g. puppet, chef, salt). I’ve also been a developer and worked closely with them on multiple teams.

        NixOS is so completely different that, when combined with having to learn the Nix language, it is quite the learning curve. I don’t see how I could responsibly convince a team to use NixOS in a professional setting from what I’ve seen. The deployment systems aren’t mature enough, the code for building system images varies, and the entire OS is very much oriented on a full adoption of everything Nix. Every other management tool out there requires you to learn some sort of DSL or code patterns, but none of them are quite as unique as Nix.

        All that said, I do like Nix and am excited for my own use of NixOS. I hope you can make some progress to make the projects more approachable or abstracted away from users to allow them to benefit. I’ve subscribed to the blog and look forward to hearing more. :)

        1. 3

          announcements about our approach to the external dependency problem

          Excited to see what the approach is!

          I recently took a pretty big swig of the Nix juice.. Quite happy with the results!

          1. 1

            Let’s see Part II - the announcement they’re going to make. If it wasn’t for that, this post would be suspect.

            There’s been a lot of discussion how Nix’s learning curve is steep and that we need to build more tools and services to bridge the gap. Get that some may think their post is off topic. And maybe it is. However, rules partly exist so exceptions can be made in extraordinary situations. And Nix is extraordinary. I’m interested in hearing of any serious group working on this full-time. Also their commentary on the hurdles from and industry perspective is welcome for me. I get most of me news on open source events here. And Nix, even as just as a protocol or implementation, is such a potential boon to open source and the world, I’ll cut the poster some slack. Groups going after Nix is very rare as industry adoption is so low. Just seeing money and full-time effort getting behind the problem is encouraging.

            1. 1

              I think nix is ready for careful industry adoption, so I’m super excited to see where you’re going with this!