Elixir Users' Survey 2016 Results

I just posted a compilation of this year’s Elixir Users’ Survey Results, if anyone’s interested :slight_smile:


That’s extremely insightful, thanks for putting it together Josh :023:

Just a quick note that we have a jobs section here on the forum as well :003:


1 Like

Also a note, ML-Flavoured Erlang is now renamed to alpaca. ^.^


“Testing is dead”
That’s… frightening.

“Open source is for communists”
Aye, comrade!

"A whopping 71% of Elixir developers rate their OTP experience as below average. This tells me that we need to do a better job as a community on OTP-related outreach. "

It also might mean that the people coming to Elixir are more mature developers and evaluate their skills more realistically. With 74% rating themselves as avg. or beginner this seams to be in line with what one would expect.
In contrast say interviewing JS devs an average person will rate their knowledge 8/10 yet a small fraction knows what a closure is :slight_smile:


Now I know what a clojure is!

Well, it might be because most Elixir developers are new and OTP until recently has been Erlang-only. It’s completely normal for most of the community to feel that way because most of the community is new (“The overwhelming majority of Elixir Users have been using the language for less than a year.”).

We have more than enough OTP resources in the form of books and videos, if not blog posts.

When I search for Clojure JavaScript I get results referring to ClojureScript, but I’m pretty sure that’s not what you’re talking about.

Good catch “closure” I need to get some sleep :slight_smile:

Check the MDN website.
It explains very well what a clojure is.

I dunno, I feel like there’s a whole bunch of magic in :sys that largely unknown to the majority of Elixir devs. Or at least I keep finding new stuff…


Very cool–thanks for doing this.

I have a minor nitpick about the interpretation of the “In what environment are you using Elixir?” question:

nearly 25% of the community is doing something non-Web as well!

Is the overlap of embedded and other nearly zero? They came in at 12% and 11%, respectively, and it seems like these were added to get ~25%. Web was 95%, so people must have been allowed to choose multiple categories. From the data provided all that’s guaranteed is at least max(12%, 11%) = 12% have a non-web application for Elixir.

1 Like

I should have made this explicit. Data is hard :slight_smile: But yes, there was extremely low overlap between “Embedded” and “Other”, so it’s (hand-wavy) probably 21% doing non-web things in total. I’d have to do a bit more work to get an exact number, but a quick glance at the data showed maybe 7 or 8 “Embedded” that also had “Other” filled in. Thanks for the question :slight_smile:


Funny, I learned about sys recently. IMO that’s the fault of people writing authoritative OTP resources—why aren’t they mentioning it? Anyway, you’re right. A lot of Erlang libraries need more exposure. Maybe this could be a good idea for a series of blog posts I can write ^^

The phrase “open source is communism” was started by Microsoft’s executive.

Microsoft’s Ballmer openly denounced open source, by saying that Linux was cancer and open source was communism.

Even Bill Gates said the same.

If there is one big company that truly opposed open source software, it was Microsoft.

That doesn’t include other terrible things Microsoft did to open source community.

1 Like

But it’s a good thing they’re starting to change, yeah? https://opensource.microsoft.com/ (193 repos so far). Not to mention the bash inclusion and whatnots. I remember reading somewhere that they are practically the current biggest single organization that contributes to open source (although them being big before the OS movement does help a lot). As much as I don’t really like MS products, you gotta give them credit for that I think.

Edit: 193 pages of repo list, not 193 repos. That’s a big number.

The deployment section’s results are interesting. Where should the Docker/Heroku responders be pointed to for better solutions (e.g., edeliver -> VPS / Dedicated server)?

And the non-web related uses of Elixir are insightful.

Javascript flavored Erlang?..perhaps that was referring to ElixirScript?

:grin: :grin: :grin: :grin: Great results, thanks for doing the survey! Elixir is phenomenal. :grin: :grin: :grin: :grin:

1 Like

The 1.4 GenServer docs include a short guide over the most useful functions from sys http://elixir-lang.org/docs/master/elixir/GenServer.html#module-debugging-with-the-sys-module


Maybe Docker/Heroku is exactly what they need?

Yes, that’s a valid response.

My request for a “better” solution is related to the following:

A whopping 71% of Elixir developers rate their OTP experience as below average. This tells me that we need to do a better job as a community on OTP-related outreach. Fault tolerance is one of the biggest selling points of Elixir, and OTP is the path to it.

I have two deployment-related takeaways from this. First, we as a community should find a way to expose new users to releases earlier in their Elixir experience. Secondly, we need more examples that clarify the simplicity and value of Erlang’s distribution – Docker and Heroku inherently cripple the value the Virtual Machine can provide.