Elixir

Ericsson had had some challenges prior to that, and they set out to fix it.

Elixir came two decades later with a different angle.

When Jose tried to figure out how to bring the power of Erlang to other communities, starting but obviously not finishing with the web.

And now there's a third language from someone who did not have a telecoms or Ruby background.

Others use typed languages and are listening to this podcast because they might want, you know, they're thinking of using the Beam.

And it's also worth mentioning to those listeners that, you know, types are in the air on the Beam right now.

You know, Elixir is growing its own set theoretic type system, which we discussed on this podcast with Guillaume Duboc and Annette Benioz a while back.

But Those are types which are very carefully being retrofitted into dynamic language.

And this is exactly the place where tools, where type checkers, where stuff like linters and ELP can help with.

Speaking of tooling, right? The big hot topic happening, especially with Elixir is the gradual type checker, right? So, but having the kind of static analysis, right, which is kind of related in there, which parts of Erlang specifically make that actually difficult to do?

Yeah, so I think this will be exactly the same between Elixir and Erlang.

And our, like the, the teamwork on the Elixir type checker, and then our teamwork on the Erlang type checker.

We do, we do kind of exchange ideas every now and again, and kind of definitely are, definitely we're watching with great excitement what, what the Elixir is doing.

And, and we're ready to steal any good ideas that, that pop up there at, at, at any time.

But yeah, there's, there's a lot of things that kind of are challenging in, in Erlang beyond that.

And then we still don't have good answer to that.

So it's an interesting one.

I don't think other projects in Elixir ecosystem, Ash has an AI policy, I think.

I don't know that Phoenix does, but it might be interesting for some folks to take a look at the nerves one.

Exactly.

I submitted a few pull requests to Elixir after I got started with using Cloud for looking at these libraries and languages.

I got to admit that in the beginning, the first pull request I made, I just had Cloud write the whole thing, but it generated a lot of text.

And then also Jose, he pointed out that first of all, I have to add the disclaimer at the top that says, you know, I use Claude to find the performance improvement that I suggested like a fix for.

This

So, Luna, I was digging into the backend of a well-known messaging app the other day, and it hit me how much of that real-time magic relies on a language most people never hear about.

I'm talking Elixir.

Elixir, huh? I've seen it pop up in job posts, but it's still a bit of a mystery to a lot of devs.

What's the big deal?

The big deal is that it's built on the Erlang virtual machine, which was designed for telecom switches in the 80s.

And that's exactly what a chat app needs, right? Everyone's connected at once.

Exactly.

And Elixir takes that rock-solid foundation and wraps it in a modern, friendly syntax.

Today, we have Roberto Loy and Michal Muscala.

Roberto started his career at Airline Solutions with me, whilst Michal and I go back from the time he was actually doing Elixir before joining WhatsApp.

And he's one of the early developers who, you know, from starting his career in Elixir in its early days, moved over to Erlang.

So welcome to the show, both of you.

Tell our listeners who you are and what you do at WhatsApp.

And that's what I'm doing at WhatsApp and Meta.

Hi, I'm Michal.

As Francesco mentioned, I started my career with Elixir, but then moved on to working with Erlang and on Erlang at WhatsApp on the same team with Roberto, focusing on developer experience, but also a bit on the virtual machine, the language, and anything that can make our engineers more productive.

We are exploring an early beta copy of a book by Lance Helverson.

It's called Functional Web Development with Elixir, OTP, and Phoenix.

A fantastic read, by the way.

Yeah.

Not to mention the concurrency issues if two things happen at once.

Right.

So to build this without a database, the author introduces us to Elixir and the Erlang virtual machine, which is known as the Beam.

The Beam is just phenomenal.

And yeah, uh, let's...

I would like to say that I worked for a long time with IoT systems, and that is the reason I discovered Elixir, actually.

And it all started that, I mean, at company, we wanted to write a VernamQ plugin.

Uh, Vern- VernamQ is basically an MQTT broker.

And we wanted to make an IoT platform as well, and we really wanted to avoid as much as possible Java, so Elixir looked like a good idea.

And so let's go back in time.

I was actually doing IoT stuff because I was doing embedded stuff.

So I've been developing before Elixir with C, C++, mostly C++.