Choosing Elixir's Phoenix to Power a Real-Time Web UI
blog.wallaroolabs.com
blog.wallaroolabs.com
Been digging back in recently and between Phoenix, Plug and Ecto it's very clear how much care and thought when into the API/architecture. Truly a joy to work with.
http://careers.wallaroolabs.com/apply/FjrSD7HjmW/DevOps-Engi...
Looking for a combination role that crosses what is often considered SysAdmin and what is considered SRE.
We don't run infrastructure ourselves but have clients who are starting to run Wallaroo in production and we are looking to hire someone who is focused on making sure the end developer lifecycle experience is awesome.
This would mean touching everything from installation of the Wallaroo development environment through documenting best practices for running in production to contributing to the core Pony and Elixir codebases in order to add new functionality.
.net - https://aspnetzero.com/
ruby - https://bullettrain.co/
php - https://spark.laravel.com/
Yes, these aren't OSS but are the equivalents for elixir (if any)?
And quite possibly there won't be something like it anytime soon, if ever. I get the impression that the Elixir 'ecosystem' is a bit more modular/manual in approach even when you compare Phoenix to Rails, which means, with all else being equal, a bit less of an out-of-the-box experience.
Of course in part this is because there are currently just fewer packages available for Phoenix. But I think it's also inherent to the 'culture'. It might never be as cookie-cutter in what it offers.
(but I'm still relatively new to the Phoenix/Elixir world, so I could be entirely wrong)
I ask because I am debating whether or not I should invest any time in Elixir, if many of the benefits of learning it are easily and cost-effectively available through AWS.
As with everything though, there are trade-offs. If going strictly AWS fits your use case, you're fine working with GraphQL, etc., go for it.
I'd suggest looking at Elixir/Phoenix if you have a problem with Rails that you think Elixir/Phoenix might be able to solve. Could be scale problems, reliability, real-time communication, etc.
I wouldn't suggest replacing your entire Rails app with Phoenix right away. If possible start with a service that may be having issues and migrate that over if possible.
The channels abstraction was the driving force behind our choice and at the time there wasn’t anything that stood out in the Java world.