![\"Blog](\"/static/7aab43bff2ee0d9b5622cee2cca893f2/1cfc2/blog-as-markdown.png\" "\\"Blog")
\n \n On renting, ownership, leverage, and why I increasingly want to own more of the things that matter.
\nThere is something satisfying about owning something.
\nNot because of the object itself. A house is still walls and maintenance. A server is still a computer burning electricity. Software is still code.
\nOwnership changes what you can do with it. Change it. Break it. Keep it ten years without asking. Make decisions that make no sense to anyone else.
\nThat last part is what I like most.
\nI’ve always built things myself when an existing solution didn’t quite fit. Sometimes smart. Sometimes a wasted weekend.
\nLast year I spent two nights on a tiny download script because I didn’t want another account and another “workspace.” Ugly script. Still runs. I still know where the files are.
\nSo this isn’t really a technology preference.
\nI want control over the things I depend on.
\nThe rent-versus-own frame isn’t original. A lot of people have said a version of this. The calls I made, and the ones I still haven’t, are mine.
\nRenting is usually the right move. You don’t need to own a car to drive, or a server to ship a site, and you don’t need to build everything because you can.
\nNeed a GPU for three hours? Rent it. Buying for a temporary spike is how you get idle hardware and a story about “investment.”
\nI learned that the long way with email. I wanted an address on my own domain. I tried doing it “properly” (self-hosted mail). Reputation, queues, TLS, spam. Owning the pipes bought me a second job, not more freedom.
\nSo I rented the infra on purpose. Cloudflare for inbound. Gmail for the inbox I already lived in. No mail-vendor bill. Custom address. Someone else’s spam war. The part I actually needed to own was the domain, not Postfix, Mailcow, or whatever else I would have been babysitting.
\nRenting the boring layer gave me more options than running a mail server ever did.
\nRent until the thing is load-bearing.
\nThe danger is renting something that becomes strategically important to you.
\nThis is the example everyone uses. I’ll keep it short.
\nRent for ten years and you bought access. Buy and some of the money becomes equity, plus a roof, tax, and money you can’t point elsewhere.
\nv/s
\nOwn it and you pick which costs are worth it. Flexibility vs control. Moving a lot > rent. Staying put > own starts looking better.
\nSame fork in software and work.
\nMost software we use isn’t ours. Email, storage, design tools, managed databases. Often correct.
\nWhy babysit Postgres when managed Postgres exists? Sometimes you shouldn’t.
\nThe sharper question: what happens if it disappears?
\nWire a few years of notes, files, and automations into one service. Pricing changes. Feature dies. Account locked. You didn’t lose an app. You lost infrastructure.
\nI dodge a small version of that with writing. Draft wherever. The durable copy of this site is Markdown in a repo I control, not posts trapped in whoever’s editor was convenient. Worse than a polished CMS some days. Harder to revoke.
\n\n \n \n\n \n \nUse services. Don’t depend on them for the irreplaceable copy.
\nI don’t have a platform audience to “own.” That’s fine. This section is about where the words live, not who claps.
\nViews and followers on someone else’s app are rented reach. I don’t play that game much. LinkedIn is a profile, no posts. The X posts live on X.
\nThe writing I care about keeping is supposed to live on the domain. A URL I control. I can open the file, change it, move hosts, keep the address. A post that only lives inside an app is gone if the account is, or buried if the feed moves on.
\nThat’s the how: not more attention. More chance the writing still exists. An email list would be a tighter loop later. Not a relationship I pretend to have.
\nA job rents your time. Good deal: pay, sales handled, less risk on you.
\nDo it for ten years with nothing outside the job and the ledger gets thin. Bills paid. Weekends gone. Experience up. Durable assets? Maybe not.
\nPay isn’t enough. Leave more valuable than you arrived.
\nIf I spend six months on a stack for someone else, I try to disclose some of that work in public. A post about the problem. A small repo. Notes I can point at later. Not the confidential bits. The craft. Same time spent. Now the work isn’t only trapped in a company’s Slack.
\nProjects can fail and still leave code or a map of what doesn’t work. Jobs don’t always leave that.
\nOnce you own a place to put things, the next thing is cheaper.
\nI already have a domain, a repo, a way to ship a page. A new post, a small tool, a note I want public: it goes in the same pile. I don’t open another account to start. That’s the trick. Not that I have a lot of posts. I don’t. The old setup holds the new work.
\nSame with the home server. Once the box is already on, the next git repo, script, an automation, or small service doesn’t need a new machine. I already paid for the electricity and the fuss.
\nSame with a skill I already paid for at a job. Next time I need it, I’m not at zero. Modest stack. Still compounding.
\nOwn too much and you become the unpaid ops team.
\nPatch the server at midnight. Full-time job, self-assigned. Not freedom. Another dependency.
\nThere’s a Fight Club line that shows up whenever I get too romantic about this:
\n\n\n“The things you own end up owning you. It’s only after you lose everything that you’re free to do anything.”
\n
First half: yes. Everything collects rent in attention.
\nSecond half: no. Burning it all down is drama, not strategy. Bankruptcy with better lighting.
\nDon’t accumulate things that own your attention without returning leverage.
\nI was trying to automate some sending. Forms, lists, little workers that should have just run. A cheap tool already did it. I still spent evenings building my own. Then I had to keep the thing alive. That’s the tax.
\nAsk: does this own me back more than it frees me?
\nI’ll rent CDNs and mail plumbing. Spend ownership budget on what compounds. Price alone is a bad scorecard. Time and dependency count.
\nYou don’t need to own the machine to keep control. You need a way out.
\nExport. Domain. Portable copy. If you can leave without losing the work, you have leverage.
\nFor anything important: what if I leave tomorrow? Can I get the files out? Do I even care about the files?
\nI could have kept everything on GitHub. Instead: bare repos on a machine I already had, SSH push, cron’d backups of the important history. Not fancy. An exit that already works.
\nI don’t mind rented tools or platform distribution. I mind no door.
\nDependence is normal. Unexamined dependence is the problem.
\nOwn these. Rent the rest.
\n\nName, domain, reputation. Existence that isn’t one app’s account row.
\n\nCopies in boring formats. If leaving erases history, you never had it. Markdown on disk is ugly and hard to revoke.
\n\nCompanies and frameworks leave. Learning and building travel with you.
\n\nPublish anywhere. Keep originals somewhere you control.
\n\nNot owning humans. A URL and a file that still exist if the feed dies. Domain first. Inbox optional.
\n\nEnough that one vendor mood swing can’t kill what matters. Not a second full-time job.
\n\nA domain, a repo, a skill you already paid for. The next thing isn’t a new account.
\nRent what changes fast, costs too much to maintain, or where someone else has absurd scale.
\nOwn what accumulates and survives a pricing or algorithm change.
\nRent convenience. Own leverage. Keep an exit door.
\nThat’s less “This is mine” and more “I decide what happens next.”
\nOwnership vs renting isn’t the fight.
\nKnowing which is which is.
","frontmatter":{"title":"The Things We Own","date":"August 11, 2026","description":"On renting, ownership, leverage, and why I increasingly want to own more of the things that matter."}}},"pageContext":{"slug":"/blog/the-things-we-own/","previous":{"fields":{"slug":"/blog/debugging-the-wifi-crash-that-froze-my-homelab/"},"frontmatter":{"title":"Debugging the Wi-Fi Crash That Froze My Homelab"}},"next":null}},"staticQueryHashes":["2276319502"]}