It is my pleasure and duty to introduce some of the companies that make Goatmire Elixir and NervesConf EU possible.

The premier Elixir consultancy. They did not create Ash Framework originally but you might know them because they sponsor a ton of the work on it and use it widely. Ash is a solution to a common consultancy tension. Customers want valuable outcomes. They don't want to be paying for the drudgework. They don't care about the login form. They do however need the login form. The moment you've built something valuable they want it filtered, paged, searchable, hooked up to an API or even an AI. Ash provides those foundations and lets Alembic spend the effort on the important parts. I see so many Elixir companies pick up Ash now and I don't think the framework has even hit it's stride yet. It is the framework that will let Elixir credibly and consistently address large organisations and enterprise use.

And Alembic is not only about Ash. I mention the Ash stuff because it is strategic, thoughtful and real investment in the Elixir community and ecosystem. Because Alembic is very much about Elixir. Their logo contains the Elixir drop if you hadn't realized.

I met Josh Price, the CEO of Alembic, in Chattanooga last year for the first time. We ended up chatting at The Moxy, The Picklebarrel, getting an accidental Smith & Wesson trucker cap in an Uber. We've established a weird tendency to walk up hills together. He is an enthusiastic Elixir developer that has embraced the fact that he runs a company big enough that he mustn't do code. I think his roots still being firmly planted in the community shows in how the company operates. Alembic shows up, they consistently offer trainings at conferences, they give talks, they contribute a lot of open source. And they have a point of view.

Alembic stepped in to sponsor as one of our early sponsors had to pull out. They bailed us out of a bad spot. And we're incredibly grateful for their sponsorship and it is par for the course in what they do. You can read their blog to get a sense of the people. If I needed a team for hire to take on big work, they'd be the first I spoke to.

Oh. And we are putting on a thing the day after Goatmire..

I have mentioned Tigris a few times in the past.

They are a special type of sponsor for Goatmire Elixir and NervesConf because they are sponsoring a project. I will announce more details when we have them. But I am very excited and it should include an absolutely ridiculous integration with Tigris' object storage.

If you are on Fly I'm sure you already use their object storage. Or you probably should.

Why would you do that?

I've gotten the feigned-surprise, put-down variant of this question a bunch of times for the type of things that I do. Sometimes it is along the lines of "why would you not use Rust instead of Elixir?" which makes it obvious that the person doesn't know the trade-offs involved or already has a strong stance on it so the question isn't real. Sometimes, for example for the 5000 Erlangs stuff it is more in line with "how is this useful?" which can be legitimate. But it assumes I'm trying to do something useful. I definitely did not start the Ampere project trying to do something useful. I was going for fun.

One asker mentioned that they'd been able to simulate a lot more connections with alot less resource usage. Definitely. So have I. It is preferable if that's what you need.

While I did the project for  fun it has proven to have a lot more utility than I expected. Being able to write tests for parts of Nerves that weren't practical to test fully. Being able to simulate the behavior of a bad network to ensure the NervesHub connection does its absolute best possible job. Essentially qemu lets us have a device in a petri-dish where we control factors from the outside. This was not what I was going for but it will be really good for us.

I think it is a mistaken assumption that creative experimentation without a plan is a mistake. And that the person doing it should be doing something different. I knew I wanted to do something to show people a lot of Nerves devices.  I knew I could do some fun NervesHub/NervesCloud demos using that and show off the hardware while I was at it. The utility in this case was accidental.

I've given many "pointless" talks that people have reached out about knowing how I did the various parts. Because the components are useful. The implementation was entertainment but it wasn't useless in the way some would think it was.

Do some stuff without too much of a plan. Ideally because you think it could be rad.

The question is good. Consider why you are doing something. For sure. But don't put a lot of weight on the exact answer all of the time.

Thanks for reading. I appreciate it.