Non-profit open source in the Danish public sector

I’ve been contracting in the public sector for several years now, and seeing how time and resources (money) are sometimes spent can make me queasy. The public sector seems immune to approaching problems with a long-term solution, no matter how brain-dead-obvious (I think) it is.

Since I work with loads of clever people, it must be the incentive structures that cause decision-makers to prioritize short-term wins over long-term sustainability… again and again and again.

I’ve been pondering how to approach the problem, and I’ve had this idea materializing for some time now. It won’t fix everything, far from it, but it doesn’t have to either. It just needs to pull in the right direction.

But first, let me set the scene.

Imagine a piece of software being used in a department in the public sector. Because other departments face the same overall problem, they also have a similar piece of software - but not the same one.

I suspect the reasons for software not being reused are manifold. But I am fairly certain that some combination of the following factors will account for the majority:

• Departments work in silos, so they rarely have any idea about what is happening in “the other silos”.
• When one department wants to collaborate with another, the others are rarely inclined to. The risk of delays and exceeding budgets are real concerns, which discourages collaboration commitment.
• Big software development consulting companies aren’t interested in suggesting reuse either. Why write a piece of software only once when you can get paid to do it twice (or more)?
• A deadline requiring a “hack” to be met on time (remember the short-sightedness?) causes department-specific implementations in the software, making it harder or even impossible to reuse the software outside the department.
• Management’s existential legitimacy is tied to the number of employees they manage. Insisting that their department is “special” and leaning into “Not invented here” amplifies the silos and leads to unnecessary rework.

So far, our team has managed to avoid falling into the “hack” trap. But all the other reasons have, to varying degrees, kept us from collaborating with other departments and reusing software otherwise well-suited for reuse.

Management simply has everything to lose when they are continuously evaluated on short-term results by their leaders.

Okay, so back to my idea.

I want to open-source the software our team has been working on, and that in itself is a huge bureaucratic challenge — not so much a technical one.

But let us, for a moment, pretend it is possible. Then I want to…

1. Create a not-for-profit organisation.
2. Provide the open-source software service at its actual cost — the actual cost isn’t just infrastructure costs. It includes everything from hardware, security scans and staff with fair wages all the way to accounting and sales meetings. But it does NOT extract dividends to shareholders.
3. Be transparent about accounting, so the “cost price” can be verified and a trust relationship can be established with customers.
4. Optionally, include a fixed markup and funnel that money directly into a fund that can be used to create other open-source software and maybe even non-profit organisations like this.

A couple of notes are worth mentioning.

First, the department is currently paying for a commercial license that I believe will become entirely unnecessary once the software is open-source. This licensing cost would more than cover infrastructure and staffing for usage significantly higher than what is required today.

Second, the public sector in Denmark seems more inclined to outsource solutions, which I think would be a perfect fit for such a not-for-profit organisation. They call it “supplier responsibility” in the public sector… It gives them someone to blame when things aren’t working, and who can blame them for wanting that? Pretty smart in theory, but who wants the responsibility of spaghetti architecture for cheapsies? Spaghetti architecture is one of the many consequences of always prioritizing short-term gains 😅

The only way a for-profit organisation can sell the same service significantly cheaper is to sell it at a loss. Sadly, selling at a loss is common when competition is expected to withdraw before you do, at which point prices can be raised to a premium, or when the actual losses are more than compensated by other services. Both practices, however, shatter the illusion of us having a transparent market.

I wish such a non-profit organisation could actually “take off”. It would benefit everyone except those who just try to extract value (money) for themselves.

One can dream 💚 (It has been almost five months since the first draft of this post.)