Build, Maintain, and Document

A note on the structures that outlast their creators.

For years, I delegated my oil changes; the upsales too. When I lived across the lot from Firestone, I let them take care of my automotive needs, completely. I walked over, handed them the keys, and returned whenever they'd call, to a freshly printed invoice, and a tuned-to-spec engine. I never bothered asking how many quarts my engine lost since I last brought it over, or whether the filter had been over-torqued. It was a system I integrated into, one of geographic proximity. Maybe that made me lazy, or comfortable, or both.

When I changed stations, that system did not follow. The alternatives were Take 5 and Midas down the strip. I let the former keep my five, and then some. They returned the favour with a gig-worker's coupon. For a Benjamin from my checking account, I was out in less than the time most pizza shops wait before offering their deliveries for free. But as I swiped my plastic, a realisation took over: I could do this myself.

So, I ruminated upon that thought, bought an extractor, the one that pulls fluids out of the dipstick. Thought I'd test the waters by doing a "quart"-erly top-off. But the extraction alone took an hour. A little under four quarts of gunk from a machine rated for 5.7 qts? A win, is a win, regardless. But I had already committed to swapping my own fluids. The capital expense on a six-month revolver was already accounted for. Why outsource, when one could keep five, and perhaps, make five?

I did not possess a drain pan, one of those SKUs oft overlooked until it is time. So, I searched for something that could perform the role. A sea green mop bucket from Target. Some 16 quarts of sheer void, wide enough to catch the splashes too. Exactly like how one imagines milking a cow. No-non-sense, straight to the pail. The jacks came out. Front wheels propped. Theory, meet praxis. I observed the jacks with precision as I crawled underneath the frame; first, the splash guard with their unique screw fittings. Then, the drain plug. An impact wrench in time, saves nine. Clearly, the extractor had not done the job. But that is a Section 179, tax-deductible, on-the-job, learning experience.

Then came the oil filter. Of course, I was not planning on doing it, given that I had only talked myself into performing a partial top-off. I bought cap-style wrenches, in advance, for D-Day—the ones with flutes. Didn't know of flutes that blew instead of being blown. Tried a strap. A glove too. Then I reddit. Googled it. Stabbed the tin can with a screwdriver. A rookie move, of course. The mop bucket held though. Called it a day.

I walked a mile to the nearest AutoZone to get the right tools for the job, first chance I got, more like a late afternoon stroll with my dog in tow. An oil filter wrench. They have an entire shelf. I might as well grab the one with the swivels as well. If it saves me the trip, on foot. And, yes, to get in the zone, I added a can of Monster to my tab. The "oil filter wrench" was enough. More than adequate. A few twists and the new filter was on, prefilled and lubricated, of course.

In sum, that oil change cost me a band, not the standard economy rate of half a Benjamin. High-mileage for 75k+ miles. Three business days of trial-and-error. One recovery day. An entire roll of shop towels. But the splash guard slid back in place. The tools stowed in crates. The engine hummed; coolant topped off too.

So I started documenting what I built, so I could maintain them; what I maintained subsequently required documentation as well. What tools I picked up, what failed, what I'd do differently. Those logs became a system. Milk crates got traded for interlocking Packouts. Linktree became host to all my links, until I swapped media, rolling out dispatches in static .html files.

So no, I didn’t pour concrete. I didn’t design a pristine workspace. I worked with what I had: a mop bucket, the wrong tools, a skeptical dog, and the internet whispering all the wrong advice. But that first oil change gave me something no service ever did — proof of concept. Once I decided I was going to go down under, I knew I had to get comfortable working under tonnes of metal. And to trust systems that begin not with knowledge — but with the decision to learn anyway. You don’t build to impress — you build to outlast. And you build to avoid calling someone for help you know may never come.