Another point worth mentioning is that Elixir is only optimised for lightweight processes. If you have to do any heavy lifting you're better off with Go or a JVM language, not Elixir. Forget math with Elixir too - it doesn't even have an arrary/vector data structure of its own. Instead it relies on Erlang's hacky implementation.

Which is exactly what many Elixir advocates say.

Contrary to what you and others try and insinuate, we are very aware that the language is not a universal tech.

So yet again, you guys are smacking at a strawman that's very easy to beat. It's quite tiring to argue with stuff that I never said (and most of Elixir's community haven't said it either). But I think I'll give up for now. :)

That's not been my experience. Whenever I have mentioned the lack of arrays/vectors in the past I have been inundated with Elixir devotees asking me why I need them when Elixir has maps with numeric keys at which point I just throw up my hands in despair.

Not denying it. Every community has bad apples.

It's not charitable to make assumptions for the entire community -- or the tech stack itself -- based on them however.

Seems like you've been chatting with the wrong people. A lot of the ElixirForum regulars -- me included -- are very practical and realistic folk who use several languages all the time.

This is putting words in the mouths of other people and is close to a flame bait. (Not to mention your gross generalization of the members of a not-small community.)

1. The Elixir community doesn't seem to aim for popularity (at least several important maintainers have said so on ElixirForum).

2. The tech is useful, scales very well and helps with the hosting costs by using less resources. That's an objective fact, proven in many blog posts where people sat down and measured.

3. Concurrency and parallelism are extremely important and will get more and more important as our CPUs can't get faster single-core performance and are just getting more cores.

It's OK if you don't care about any of those things. Really. I personally never got offended if something I like is disliked by others (as we should all).

But making a strawman and then crushing it is not a productive way to discuss either.

> The Elixir community doesn't seem to aim for popularity (at least several important maintainers have said so on ElixirForum).

Wow, then why does every Rails thread has an Elixir guy trying to tell us what an inferior software we're running? For the sake of humanity?

> and will get more and more important as our CPUs can't get faster single-core performance and are just getting more cores.

I don't think it will become more important in web dev world than it is now. We're here churning out back office CRUD apps like crazy, I don't know what you're doing. For a small team trying to get something up and running this really doesn't matter and that's what Rails is optimised for. If scale ever matters, if you scale that big, you probably already have the resources to deal with it. For the few startups that have huge scale + have very scarce resources and Elixir works well for them - great! I have no issues with that at all.

> Wow, then why does every Rails thread has an Elixir guy trying to tell us what an inferior software we're running? For the sake of humanity?

If you prefer that melodramatic phrase. I've been a Rails dev for 6 years and I know that many current Rails devs don't dare try something else -- call it a survivorship bias, fear to abandon what pays the bills now, habits, and most of the time you won't be wrong.

Every now and then I speak with some of them and we exchange ideas.

And you should stop stereotyping, it's juvenile and definitely has no place in HN.

> I know that many current Rails devs don't dare try something else

OK, and I'm the one generalising yes? Clearly you have zero bias towards the Rails devs.

How about simply becoming really good in a skill instead of jumping stacks every 5 years? To REALLY become good in Rails, to the point where you can write gems for activerecord for example that do something cool with arel, or follow the work Tenderlove is doing easily, this type of good takes effort. and time. Some people don't care at all and will happily switch and rebuild that expertise (or will never become experts in anything). That's totally fine. I value expertise, and won't make the switch for something that on the surface doesn't look like will benefit me at all (again, concurrency as it is in Rails is just fine for my career path and I think most Rails devs are on the same boat).

> is close to a flame bait. (Not to mention your gross generalization of the members of a not-small community.)

You're the one who chose to use the term cargo cult a few comments above

Well, I've been to Rails meetups. I can bet a few bucks that you haven't been to Elixir ones.

No I haven't. I'm not on a constant chase to find the better tech. I wanna write good code and come back in the end of the day to eat dinner with my wife. Even if Elixir is objectively better for web development than Rails (which I highly doubt), I wouldn't necessarily want to invest the effort in learning it. What's so interesting in learning 10 different ways of writing yet another CRUD app? Some people get a kick out of it. These people don't have kids lol.