The Spiral
It started out as a simple desire to build a software project to help out a friend. Several weeks in, I have a laundry list of tasks and very little product to show for it.
I was a "web site designer" for approximately 10 years as a freelance/sole proprietor.
In that time I did a lot of interesting work. I had clients on 4 continents. I worked with Ricki Lake's production company on the site for the release of her 2008 film The Business of Being Born. I partnered with massive, nationally recognized design houses. I built launch sites for several startups. I did work for multiple UK government agency campaigns. Ultimately, I don't think I was very good at it as the income was too volatile to sustain my family.
But I'm older, wiser and better resourced, both in social terms and financially. The one thing I do not have more of is time. Having decided that I am not busy enough in my day to day life, I got the idea that I should try to rebuild my old and dusty frontend skills. I went in search of a project, and landed on an idea I had with a friend to build a Shopfiy app to help him out with his business.
Now, I used to do Shopify development, but this was like 10 years ago. Web sites were still mostly written in PHP or some RoR. Web frameworks like angular and react were very much in their infancy at this time. I very rarely write frontend code at all anymore, much less stuff for interfacing with "app stores" because my work now is mostly data engineering and API building, with a healthy dose of infrastructure as code.
Shopify now more or less requires you to write apps as React apps - that is, all of their provided tooling and tutorials are in support of a React/Node framework, using their own designed components CLI tools. In my younger days I would have bristled at the idea of having to use someone's chosen stack. But now, I said "bet."
Of course, in order to launch a Shopfiy app I would need a few other things to fall into place. In my professional life I use a pretty hefty stack of enterprise tooling to get my day-to-day work done. Many of these pieces I can do with out, but some of them I've had to find replacements for. Since I'm not yet making any money off of this hobby the objective is to do them on the cheap. But doing infrastructure services on the cheap means the investment of time.
And so I find myself with a long list of tasks:
- Deploy a homelab setup where I can run my own private infrastructure
- Set up some core services for source management, CI, project and task tracking, design work etc
- Build a landing page and support site for the Shopfiy App
Those things done I can move on to more administrative tasks:
- Set up a merchant account
- Get alternate email, phone and other contact systems set up
- Do research and advance marketing
At every one of these tasks, I come up with 5-10 more new things to do. The to- do list is getting seriously long. In terms of designing and building the actual app you know, the reason I started doing this to begin with, I've made shockingly little progress.
On the upside, I am building infrastructure that I can continue to reuse in future projects, meaning this process will only get faster. Building the landing page meant a fast crash course in the simplest elements of React, and I am feeling confident and productive with Next.js - a React based app framework from Vercel that I chose to get my React chops up and running in advance of App development.
And, I can see the edges of my projects coming together. In my previous life I would have skipped most of these steps in an attempt to speed-run victory. Instead I'm working methodically and cautiously to build all the scaffolding to support the success of the thign I'm really trying to build. And ultimately it's very satisfying. I'm leveraging skills I haven't used in over a decade and learning new ones.
So while it feels like my tasks are spiraling out of control, I'm actually steadily closing in on my goal. I'm able to pick the "most important thing" each day and make some measurable progress on it, even if I only have an hour or two. It's very satisfying to see this coming together and I can't wait to see how fast thigns start coming together when I get to actual project work.
See what's coming at Pixel T Software and hopefully soon, Verdalink