The year after I graduated with my CS degree, I tore up my backyard.
Not literally. I built it up. Five raised beds, filled with quality delivered soil, drip lines run to every one. Trellises. A seed-starting setup that took over the upstairs office by February. I'd discovered Square Foot Gardening that year, and I went in the way a newly graduated software engineer goes in on anything that combines systems thinking with something physical: completely.
I also wanted to track everything.
That's not a quirk if you know what a CS degree does to your brain. I wanted to know what I planted, what variety, how many starts, when I put them in the ground, when I harvested, how much, and how it tasted. I wanted to look back at a season and understand why a bed underperformed, not guess. I wanted a seed catalogue I could use for real planning, companion planting guidance built into the layout, fertilizer logs tied to actual performance data.
So I went looking for an app that could hold all of this.
The best I found was GrowVeg. It's a good tool, genuinely good at layout and rotation, clearly built by people who care about gardening. But it topped out right where I needed it to keep going. No meaningful analytics layer. No harvest tracking with real depth. No way to connect what I put into a bed with what I got out of it. I kept a spreadsheet for everything it couldn't do, which started to feel like the point.
Around this same time, I met another serious gardener. This person was deep into soil health, really deep: the kind of deep where you're sending samples off to a lab and waiting on results. They had built a spreadsheet of their own. You entered your bed dimensions and your nutrient readings, and it calculated exactly how much of each amendment to apply: lime, sulfur, gypsum, greensand, down to the pound.
I looked at that spreadsheet and thought: yes. That too. That's part of the app I want.
So in 2010, I started building it.
I made real progress. Real enough that the shape of the thing was visible: the seed catalogue, the layout canvas, the tracking layer. Then I ran out of steam and moved on to another idea. Every developer who has ever started a side project knows exactly how this goes, and I'm not going to pretend my experience was different. The idea was right. The timing wasn't.
The gardening continued anyway.
Seasons passed. I moved to a new house. Different beds, different orientation, a frustrating sunlight situation in the back that has me slowly planning to eventually take over the front yard instead. Every spring I started planning, and every spring I reached for the tool that still didn't exist. GrowVeg was still around. Planter had come along and built probably the best mobile experience in the category. Neither of them had closed the gap. Serious gardeners were still doing what that other gardener had been doing: running nutrient math in a spreadsheet, tracking harvest data in a notebook, keeping fifteen years of variety preferences in their heads. Not because they didn't want better software. Because no one had built it.
Vernal is that 2010 project, finally getting built properly.
The tracking layer is the point of it. Harvest records tied to specific beds, yield comparisons across seasons, soil and amendment logs: the data serious gardeners are already keeping somewhere, scattered across spreadsheets and notebooks, finally in one place and connected to everything else. What that spreadsheet was doing for nutrient calculations, Vernal does across your whole garden history.
The rest is built to support that. A layout canvas for standard beds and irregular ones that don't fit a grid. A seed catalogue that works the way a physical one does: each seed rated, annotated, marked as a favorite, pulled into your layout when you're planning a season. Growth visualizations at every stage from seed to harvest-ready.
It's a web app, progressive and mobile-friendly. The core is open source. Your data is yours: exportable, readable, never locked behind a subscription. That last part matters to me as a gardener for the same reason it would matter to you.
I'm building this solo, in the open, and this devlog is where I'll document it. No cadence promises. I've been around long enough to know how that ends. But there's a lot to cover: design decisions, the data model, the parts that turned out to be harder than expected and the parts that came together faster than they had any right to. The interesting stuff.
The waitlist is open at vernal.garden if you want to be there when it launches.