When comparing F# vs Elixir, the Slant community recommends Elixir for most people. In the question“What is the best programming language to learn first?” Elixir is ranked 9th while F# is ranked 30th. The most important reason people chose Elixir is:
Leverages the existing Erlang BEAM VM
Specs
Ranked in these QuestionsQuestion Ranking
Pros
Pro Concise syntax
F#'s syntax tends to be terse while remaining very readable and easy to understand without being a chore to write.
Pro Easier transition from other paradigms
Since F# is not a purely functional language, it lends itself to being more easily picked up by programmers that have experience with other paradigms.
Pro .NET Interoperability
Since F# runs on the Common Language Runtime or CLR, it has access to the entire .NET Framework, as well as libraries written in other .NET languages such as C#, VB.NET, and C++/CLI.
Pro Multiplatform, it runs on .Net Core
Forget Xamarin and Mono. F# now runs on the multiplatform .Net Core!
Pro You can run F# in the browser
Thanks to a project like Fable, you can write and run your client app in F# and run it over JavaScript. It allows you to share code between your server and client.
Pro Natively supported by Visual Studio
This is a .Net language natively supported by Visual Studio. Though it is not as tooled up as C# the support is still substantial. In particular, C# deployment scenarios can be enabled for it with small C# wrapper projects. Integration with Visual Studio provides: IntelliSense, debugging, projects an other features.
Pro Fall into the pit of success
F# directs you into a workflow where the right way is the path of least resistance.
Coming from a C# background, its restrictions might feel arbitrary at first (e.g., what do you mean my code has to be in dependency order!? Arbitrary alphabetical or bust!), but you'll soon realize that your code is cohesive, concise and consistent in a way that it never was before - and you can compile and run with confidence!
Pro F# is supported by Xamarin
Thanks to Xamarin, F# is a functional language that you can use to build for iOS, Android and Windows.
Pro Great for concurrency
Leverages the existing Erlang BEAM VM
Pro Great getting started tutorials
The tutorials are very clear and concise (even for a person not used to functional programming). Plus they are also very mobile friendly.
Pro Powerful metaprogramming
Write code that writes code with Elixir macros. Macros make metaprogramming possible and define the language itself.
Pro Full access to Erlang functions
You can call Erlang functions directly without any overhead: https://elixir-lang.org/getting-started/erlang-libraries.html
Pro Scalability
Elixir programming is ideal for applications that have many users or are actively growing their audience. Elixir can easily cope with much traffic without extra costs for additional servers.
More details can be found here.
Pro Great as a first functional programming language!
Pro Great documentation
Elixir's documentation is very good. It covers everything and always helps solving any problem you may have. It's also always available from the terminal.
Pro Syntax is similar to Ruby, making it familiar for people used to OOP
All of the benefits of Erlang; without as steep a learning curve of prolog based syntax. Elixir is heavily inspired by Ruby's syntax which many people love.
Pro Easy to download libraries
Comes with built in build tool called "mix". This will automatically download libraries and put them in the scope of the application when you add them to the "deps" function and run mix deps.get
Cons
Con Has no ad-hoc polymorphism (á la Type Classes)
You have generics, you have interfaces, you have inheritance, you have a lot of things at your disposal but you don't have Type Clases. They can be emulated using some clever constructs but there's nothing like having the real thing.
Con Infested with OOP
Still relying on .NET OOP libraries.
Con Deployment is still not as easy as it should be
Con Some design choices may seem strange
Some design choices could have been a little more appealing, for example: using "do...end" comes natural in Ruby for blocks but Elixir uses them for everything and it looks pretty weird:
Enum.map [1, 2, 3], fn(x) -> x * 2 end
or
receive do
{:hello, msg} -> msg
{:world, msg} -> "won't match"
end