Sometime in early 2019 I started to have that itch. That feeling that every creator has when they regularly look at something they made. I felt that itch to build a new version of my site.
I built the first version of my site as a proof of concept. I wanted to learn how to use Gatsby, and since thetrevorharmon.com had been sitting vacant for a few years, it felt like a natural fit. I challenged myself to release something within a week's time. It didn't need to be great, but it needed to be publicly accessible. I cobbled together a site with some hastily written content, a first-draft design, and some mediocre code to go along with it.
By July 2019, I could no longer ignore that itch and started in on a redesign. I made a Github issue and a laundry list of all the things I wanted to change. I started on a design sometime in late September, and started development in early November. I got to a stopping point in mid December and stepped away for a few weeks for a holiday break.
About my process
I plan to go into greater detail on my design and development process in future posts, but here are some high level points about this redesign:
- The entire process took about 6 months. The first half was spent exploring and deciding what I wanted, with most of the work happening in the second half.
- The pull request to implement the redesign includes 4,533 new lines of code and removes 3,607 lines. It was almost a complete rewrite of my site.
- Not included in the previous statistic is the addition of dark mode to the site that happened in November 2019. That was 1,433 new lines with 768 lines removed. That was a foundational piece that I decided to release separately.
- I made the conscious effort to care more about accessibility as part of this redesign. As a self-taught web developer, accessibility unfortunately tends to be later on the list of things to learn (and I'm finally further down the list). With this redesign, I improved the Lighthouse accessibility score from 89 to 97. I'm not "done" with accessibility, but moving in the right direction.
- Until now, I've only used fonts available on Google Fonts. I decided to use Inter, and it's even a variable font! I have a post in the works that goes in depth about how I prepared and implemented the font.
What did I learn from this process?
- It was really tempting to throw away my old repo and start from the ground up, but I instead decided to take an iterative approach. While difficult at times, I found that this kept me moving foward. I think it would have been more difficult to start over, because of how intimidating it would have been to start from zero.
- It's worth the time to focus on design and development separately. Before this I had tried to do both at the same time, but focusing on them separately kept me from getting bogged down in the code too early.
- Even for a small site like this one, redesigns are a lot of work!
The last thing to share is the code itself!
Sharing my work
The most exciting thing for me is that I have decided to make the repo for this site a public repo. Shawn Wang has a great "golden rule" of learning in public, and I want to do a better job of that. So if you want to see how I'm learning and see the work in progress that this site is, feel free to poke around the Github repo.
I'm also keeping up the old design of this site at previous.thetrevorharmon.com (for a few months), and I would love to hear what you think about how this one compares. In what ways is this version better? Or worse? I'd love to hear what you think.