I’ve never met a homelab that was “done.” Mine certainly isn’t — Moya gets a new workload, Talyn gets torn down and rebuilt to test an idea, Pilot’s storage layout gets reorganized for the third time this year because I read one blog post about ZFS pool topology and suddenly the old layout looked wrong. If you run a homelab, you already know this feeling. There’s always one more thing.
I started reading into why that is, less as a personal diagnosis and more out of the same curiosity that drives everything else on this site — how does that work? — and it turns out there’s a well-documented pattern in ADHD research that describes the homelab condition almost too precisely.
The Novelty Engine
ADHD brains are consistently described in the research as novelty-seeking, driven by lower baseline dopamine and a corresponding pull toward anything new, stimulating, or unsolved. The beginning of a project is maximally novel, and interest naturally wanes as work becomes familiar. A new Kubernetes operator, an unfamiliar storage backend, a different GitOps pattern — these are basically novelty generators with a CLI.
That’s not a knock on homelabbers generally or an attempt to diagnose anyone reading this. It’s just a strikingly good description of why a hobby built entirely around “spin up something new and see how it behaves” is so magnetic for a certain kind of brain, ADHD or not. Kubernetes in particular is a bottomless well of new things to learn — operators, CRDs, service meshes, storage classes, scheduling policies — which makes it either the best or worst hobby to hand someone whose brain is wired to chase the next unfamiliar thing.
Hyperfocus Is Not the Same as Discipline
There’s a popular myth that hyperfocus is just intense willpower aimed at a task. The research says otherwise: hyperfocus tends to be triggered by novelty, intrinsic interest, and emotional activation — not by external deadlines or rational decisions to work. You don’t choose to hyperfocus on debugging an ArgoCD sync failure at 1am. It just happens, the same way it might happen to someone playing a video game instead.
That distinction matters because it reframes a pattern a lot of homelabbers recognize: the weekend where you rebuilt an entire cluster from scratch in six uninterrupted hours, followed by two weeks where you didn’t touch the lab at all. That’s not inconsistency or laziness. It’s a documented cycle sometimes called the “hyperfixation hangover” — an abrupt loss of interest following a period of intense focus, often tied to dopamine depletion and the novelty wearing off.
The Project Graveyard, Homelab Edition
Every homelab has one. Mine has a half-finished Prometheus alerting setup that’s been “almost done” since February, and a Talyn experiment with a service mesh I never actually got working end to end. Across ADHD communities, this gets a name: the project graveyard. It’s described as practically a rite of passage rather than a personal failing — evidence of a mind that sees possibilities everywhere, not a monument to inadequacy.
What’s notable is why projects stall specifically at the end rather than the beginning. Completing complex work requires sustained effort on tasks that may no longer feel interesting — testing, documentation, the unglamorous finishing details — and that demand on executive function arrives exactly when dopamine-driven interest has already declined. Translated to homelab terms: spinning up a new service is the fun part. Writing the runbook, setting up monitoring, and documenting the failure modes is the part that actually finishes the project, and it’s also the part your brain is least equipped to want to do by then.
Reframing, Not Excusing
I’m not writing this to hand out a free pass for unfinished YAML. The research is also pretty clear that this pattern can lead to unfinished projects and burnout regardless of whether it’s also a source of creativity and genuine engagement. Both things are true at once.
What I do think is worth sitting with: if your homelab looks like a graveyard of half-finished clusters and abandoned Helm charts, that’s an extremely normal pattern with a real explanation behind it, not evidence you’re bad at this hobby. The next post is the practical half of this — what the research actually suggests for working with that pattern instead of constantly fighting it, applied specifically to Kubernetes and homelab projects.