Ash and "business value"

No images? Click here

Oredev 2026 line-up (I am in it!)

This big ol' swedish dev conference has announced its line-up and I'll have the pleasure of returning as a speaker. Which is convenient since I was already going to be there. Doing .. whatever it is I do. Video below for my prior exploits.

I will be speaking on the stuff other runtimes can't do. Meaning I'll be showing off the BEAM, from Elixir. Probably doing a bit of live tracing of NervesCloud and so on. If you have favorite stuff the BEAM can do in DevOps, observability and delivery aspect, feel free to respond to this email and contribute it to my raw materials for the talk.

I built a registry for the EEF

The Erlang Ecosystem Foundation has a problem. Well we have many, that's why exist, to take on problems for the ecosystem but we have one major one that will be on the list to solve forever. But we need to chip away at it.

Lots of companies use the ecosystem. Few companies contribute to the ecosystem.

The first contribution a company can and should make, entirely free and still very valuable is to admit to using the ecosystem. So we offer a registry in which companies can sign up, verify a contact and show off their logo.

Sign up your company or find someone who can. Please and thank you.

I'll expand on the how and why further down as today's editorial piece.

 

Goatmire approaches.

Teaser video for Goatmire Elixir 2026
 

Building the EEF Registry

The technical bits are not particular exciting which is a sign that they are pretty well put together as far as building blocks go. The system is all Ash. Why? Because I do not have a lot of time to make decisions about every little part of what is at its core a CRUD app.

Ash, Cinder data tables, Postgres. That's about it. It exposes some REST APIs for staff convenience and automation. It has a mailer for auth and logins. Common stuff. I don't believe I have added any MCP stuff via ash_ai. I generally do that for my own internal tools right now because I find it convenient and useful. Especially for this type of business software. Having natural language input to a systems like this is super convenient. I do that for the Goatmire sponsor CRM. I don't do it for this. Most of the good information is not entered by us. Importantly it is self-reported by the companies which makes it more likely to be accurate.

So how does the registry work and why does it matter?

The core of the design is that we want a low-low-low friction way for companies to contribute. In sales terms this is top of funnel stuff. It is the free part, the start of the relationship. I am confident that there are more companies that no EEF member has ever heard of that use the ecosystem than the ones we, in total aggregate, know of. The unknowns are massive. I go to a random town in Sweden large enough to have a major burger franchise and there is at least one Elixir company. If we assume one spawns at every McDonalds there should be 40.000 companies out there but I've no idea.

And we won't reach all of them. But the ones we do reach we can't just go. "Hey, drop 2000 Euro in our hand right this minute." That's not a relationship, it is not a clear exchange of value. There is no there there. It is begging and non-sustainable.

The goal of registration is to open the communication channel and start to build the relationship. This is also why the company self-registering is important. It is an opt-in, it is consent and it is the company offering a little trust to us that we won't abuse their willingness to engage ever so slightly. Then it is on us to nurture the relationship.

This is why you can't register another company. It wouldn't pass our approval process. To absorb any excessive helpfulness we do have a tip form which feeds into our outreach bucket. It becomes raw materials for finding new companies. Also useful but mostly a bonus and a sink for a certain overzealous type of helpful energy that occasionally shows up in work like this.

Currently the offering and the mechanism for building the relationship will be quarterly email newsletters which will cover the major EEF initiatives, probably an overview of the events happening over the year, strategic help and guidance. The EEF is working a lot on security and supply chain. Driven by legislation in many territories that expect more from companies than what the current norm is. This is not about the latest whiz-bang library. More strategic, more big picture. And it shows the work of the EEF. I imagine the first one will need to go relatively deep on the CVE-generating efforts of Peter Ullrich and how Jonatan Männchen is only one man processing all of this. Valuable efforts are happening continuously and making them known to companies is a much better basis to recommend sponsorship. The limited cadence should ensure it is not noisy to busy technical leadership.

Of course we'll suggest that companies sponsor these efforts. There is work underway to overhaul the sponsorship aspect of the EEF as well to make it a lot nicer and more punchy. But the expectation is that it will take time for companies to become interested in all this. The emails are a heartbeat and free helpful strategic intelligence. Take it or leave it.

Over time we want the registry to be a double-sided marketplace of sorts. The EEF is structured to be a neutral non-profit entity for the ecosystem and as such companies can feel safe registering with us and our members can feel safe with us. And we can mediate exchanges that they might not be comfortable doing elsewhere. There are many cases where actors in the ecosystem want to reach companies. Events want sponsors. Consultancies, contractors and freelancers want to reach B2B and B2C companies. We can find ways to mediate those connections, creating useful exchanges in the ecosystem and charging for some of it to support the work of the foundation. Companies want to reach the ecosystem with things like job postings but they want a filter on the amount of applicants they'll get. The membership of the EEF is a very high-quality filter signal.

These exchanges are not in place. They all need to pass through approvals within the Sponsorship working group and some will certainly need to rise to the board level.

One thing I love about this scheme, because it is all a scheme, carefully laid out plots and plans in service of the ecosystem. Marketing and sales have an aspect of trickery to them. In this case I'm quite confident that the magician's trick will, over time, simply delight the audience. Companies register as a duty, a contribution, a nod or a thank you. But we'll leverage the mass of companies in applying for grants and proving ecosystem size which hopefully scores more funding. Which allows doing better work for the ecosystem which should encourage more sponsorships. The idea is to create a very valuable list of companies and a very valuable set of members in the foundation. And mediate the connections to the benefit of both. We don't give your information to any company. You don't get the information to contact any company. But we can help you achieve useful connections.

We'll see how much of it pans out. We have already learned about new companies by just sharing the registry. Please help share the signup to BEAM companies you know. And let me know if you have questions or thoughts about the plans. Nothing is set in stone. There are many things we could potentially do. Very open to any criticism as well. I'm being very transparent about motivations here and I don't know if it will come across as cynical or crude. It is business, it is sales. I fundamentally believe that good sales are all about establishing trust and creating a mutually beneficial exchange. That's the goal.

I call the project itself `jewel`. In the hopes that the registry we build will be the crown jewels of the EEF.

 

Have a great weekend, thanks for reading.

 
 

This is an email from Underjord, a swedish consultancy run by Lars Wikman.

Everything else is found at underjord.io

You signed up for this newsletter and confirmed the subscription. If you want to stop receiving it. Just use the link below.

Preferences  |  Unsubscribe