← All writing

What should be shared

The center has a job. So do the edges.

Three distinct silver grain forms rise from one continuous field against black

Put several different businesses under one name and a question appears almost immediately: which decisions belong in the middle?

The tempting answer is more of them. A common name, common process, common design language, common way to launch. It looks efficient on a diagram. In practice, it can leave every company sounding like a department of the same company.

Einhar is being built around a different bargain. The center should take on work that becomes more useful when it is shared. Each company should have room to decide what it makes, how it speaks, and what kind of relationship it has with the people who use it.

The work worth carrying together

Some problems are expensive to solve from zero each time. How do we investigate an idea before committing to it? How do we build software that can be maintained? How do we record a decision so the next person can understand it? Research habits, engineering knowledge, and the unglamorous work of keeping things running can improve across projects.

That doesn't mean making every team use the same answer. A shared foundation is more like a workshop than a mold. Tools and experience are there when they help; they should be questioned when they don't.

Consider the distance between a creative tool and a game studio. They might both benefit from good technical infrastructure and careful user research. They will still need different rhythms, different ways of judging quality, and different public voices. If the parent company makes those differences harder to express, it has taken on the wrong work.

A name has to be earned

We want the companies we create to have identities of their own. That is not just a branding choice. A product name should tell people where to direct their attention and what to expect. It should belong to the work behind it, rather than being attached early because the organization chart needs another box.

The same restraint applies to what we call a company. An idea can be promising without being a business yet. A prototype can be useful without needing a grand introduction. We would rather let a project become specific than ask it to perform certainty before the evidence is there.

The clock matters

Long-term ownership gives this structure a reason to exist. If we intend to stay responsible for what we make, durability and trust stop being abstract virtues. They affect the ordinary choices: what to maintain, what to simplify, which opportunities to pass on, and when a product needs another round of work.

Ownership doesn't guarantee that every experiment will last. It does give us a reason to be honest about what is working and what isn't. This Journal is one place to keep that record in public: the questions, the decisions, and, when we have them, the results.

That is the test for Einhar's shared foundation. It should make the next good decision easier without making every company look like the last one.

All writing