./writing/gearious/origin
Fourteen Years to a Wireframe
In 2011, I was a serious backpacker. Not casually serious. Seriously serious. The kind of serious where you weigh every piece of gear you own on a kitchen scale, log it into a spreadsheet, and then quietly judge the long-distance crowd cutting the handles off their toothbrushes for being so committed to weight savings — while also, simultaneously, wondering whether you should be cutting the handle off your toothbrush. (Reader, I did not. There are limits.)
I owned five different cooking stoves. The lightest titanium burner for long miles where every ounce mattered. The most convenient pocket-rocket for an overnight when I wanted to be making coffee within ninety seconds of unzipping the tent. Three more for trips of intermediate length and varying levels of forecasted rain. If you have ever owned five cooking stoves, you don't need me to explain. If you haven't, no amount of explaining is going to help.
The spreadsheet did its job. What it couldn't do was let me try alternatives quickly, share my actual setup with other backpackers in a form anyone could verify, or show me what swapping the titanium burner for the pocket-rocket would actually do to my total pack weight without me re-typing the whole list. I started sketching what that tool would look like instead. Late evenings in the conference room of the agency I was working at, filling a wall-to-wall whiteboard with hundreds of sketches — the interface, the workflows, the integrations, the partnerships, the question of how something like this could ever make money. A virtual closet of every piece of gear I owned. A virtual backpack I could load and re-load with a few clicks. A way to show my receipts.
I pitched it to the owner of the company I worked for. The idea didn't fit the business, and I didn't have the engineering depth to build it on my own — not at the scale it needed to be built. The sketches went into a folder. The folder went on a shelf. The five stoves stayed in a closet, which, looking back, is roughly the same outcome.
the folder on the shelf
Every few years I'd run across the folder. The idea still held up. My skills had grown — I'd taught myself a lot of front-end, taken a Ruby on Rails class back when Rails was the thing every designer was learning for some reason, built a small Instagram-style app to prove to myself I could — but the gap between what I could design and what I could ship on my own was still too wide. I'd open the folder, page through the sketches, recognize the shape of an idea I still believed in, and put the folder back on the shelf.
This cycle ran for about a decade. Three or four times I picked up the folder. Three or four times I put it down. The folder is, at minimum, a very patient folder.
the part I didn't expect to write
The honest version of what happened next has to start with the fact that I was, for a long time, openly skeptical of the AI tools that were starting to land. I like the craft. I like aligning every pixel. I like thinking past a single feature to the shape of an entire user journey, and I was not convinced that a tool trained on the internet was going to do that work the way I wanted it done. I held that line for a while. Some of it I still hold.
What changed wasn't a conversion. It was a recalibration. The tools kept improving, the people around me kept finding real uses for them, and at some point the question I'd been asking — can these tools replace what I do? — gave way to a more useful one: what can these tools do that lets me work on things I couldn't work on before? That's a different question. It doesn't ask the tools to be something they aren't. It asks where the leverage actually is.
I started small. Drafting emails I didn't want to write. Asking it to turn one of my unfiltered Slack rants into something a coworker could read without flinching. Feeding it a Figma screen or two and watching it generate a working version of something I'd designed — close enough to be useful, not close enough to ship without me. The dabbling phase. I wasn't building anything serious with the tools; I was learning the shape of what they could and couldn't do.
The realization came on at an angle I didn't see coming. I'd spent fourteen years assuming I needed to become a full-stack developer to build Gearious. What I was watching, in the dabbling, was the size of that gap changing. The thing I'd been waiting on — for my engineering skills to catch up to the scope of the idea — wasn't the only path anymore. I could close the gap from the other side. I didn't need to learn enough to build the whole stack from scratch. I needed to learn enough to be the designer of a product whose code I could partner with a model to write.
I would like to say I had this thought once and immediately acted on it. I had it many times and circled it cautiously, the way you might circle a hornet's nest you're not yet sure is empty.
one thing fourteen years changed
There was one thing the long gap had changed about the project itself. I'd slowly drifted out of backpacking — not because the adventure had lost its pull, but because the appeal of covering more ground had moved me into adventure motorcycling. Same instinct, different vehicle. The same packing problem, scaled up: more bags instead of one, weight still mattering, and a new constraint that the backpacking version never had — balance. Too much weight on one side of an adventure bike makes it handle poorly, which makes it more likely to go down, which means picking it up out of a ditch with everything you packed still strapped to it. This is a problem with strong personal motivation attached.
The pivot from backpacking to adventure motorcycling was the smallest design decision in the whole project. Same gear-library model. Same trip-planning workflow. Same instinct to make the invisible visible. One new headline feature — a real-time left-right balance indicator — and the rest of the architecture carried straight across. Fourteen years of sketches that had been built for a backpacker's problem turned out to be ninety percent built for an adventure rider's problem too. That's the kind of thing you only find out by waiting fourteen years, which is not a methodology I'd recommend, but appears to have worked.
the wireframe that brought it back
Eventually I picked up ChatGPT properly. Described the idea to it the way I'd describe it to a designer I was bringing onto a project — what it was for, who it was for, what the core experience was. It generated a black-and-white clickable wireframe. Not the wireframe I'd have made. A wireframe.
And that was the moment. Not because the output was good, exactly. Because the output existed. Fourteen years of sketches I'd never been able to make real, and a tool I'd been side-eyeing for years had handed me a clickable artifact in an afternoon. The folder came off the shelf for the last time.
The wireframe wasn't a product. It was a verdict. These tools have potential. That verdict was enough to send me off to learn more — properly, this time, not as dabbling — and the work that followed is what the next post is about.