I have been watching Elixir YouTube videos pretty much every day for the last few weeks. I guess there was an Elixir conference recently and after I watched a couple, YouTube has been sending me a consistent stream of Elixir content.
I really want to try out this language. I love the idea of Erlang but the few times I've had to deal with it (an ejabberd chat server was one) I found it to be a bit too quirky. Every video I've seen with Joe Armstrong leads me to believe he was an under-appreciated genius. Isolated processes and message passing (Actors) seem a good candidate for the future of distributed programming (see this recent video from Crockford on his new Misty programming language [1] - notice how the questions were about Elixir/Erlang). I love pattern matching and elixir uses this to a high degree.
However, I am not 100% sold. Clearly this community punches above its weight, and the BEAM is obviously no slouch and is a brilliant choice for a platform to build upon. But I get concerned looking at the magic of LiveView (and LiveBook). My experience is that if something sounds too good to be true, it often is. A lot of the jargon is also totally unfamiliar to me - things like GenServers and other esoteric kinds of platform specific knowledge. It feels like stepping into a one-way portal, I'm not sure any knowledge in that universe is transferable.
There is a huge risk in niche languages and platforms. Hard to hire for, hard to find libraries for, hard to find blogs/tutorials/etc. With alternatives like Go having a pretty respectable concurrency story ... it is pretty hard to justify taking the risk on such a small community.
But kudos to that community. It is honestly impressive to me what they have achieved. I'm also very excited to see what happens with their type system implementation.
1. https://www.youtube.com/watch?v=R2idkNdKqpQ
I really want to try out this language. I love the idea of Erlang but the few times I've had to deal with it (an ejabberd chat server was one) I found it to be a bit too quirky. Every video I've seen with Joe Armstrong leads me to believe he was an under-appreciated genius. Isolated processes and message passing (Actors) seem a good candidate for the future of distributed programming (see this recent video from Crockford on his new Misty programming language [1] - notice how the questions were about Elixir/Erlang). I love pattern matching and elixir uses this to a high degree.
However, I am not 100% sold. Clearly this community punches above its weight, and the BEAM is obviously no slouch and is a brilliant choice for a platform to build upon. But I get concerned looking at the magic of LiveView (and LiveBook). My experience is that if something sounds too good to be true, it often is. A lot of the jargon is also totally unfamiliar to me - things like GenServers and other esoteric kinds of platform specific knowledge. It feels like stepping into a one-way portal, I'm not sure any knowledge in that universe is transferable.
There is a huge risk in niche languages and platforms. Hard to hire for, hard to find libraries for, hard to find blogs/tutorials/etc. With alternatives like Go having a pretty respectable concurrency story ... it is pretty hard to justify taking the risk on such a small community.
But kudos to that community. It is honestly impressive to me what they have achieved. I'm also very excited to see what happens with their type system implementation.
1. https://www.youtube.com/watch?v=R2idkNdKqpQ