Building My Site Part I: Decisions
Making decisions can be an arduous task. With the blitz of build-tools and boilerplates it can be surprisingly complex to arrive at a well-informed selection of technologies. It seems like the more you know, the less you do.
With that, here’s what I ended up with, and why.
Some History #
Before all this I had a self-hosted version of Ghost. It’s a deadly platform and incredibly easy to get up and running with. After a while I decided to move to a hosted version with them when I had less time for maintenance. It was less for me to think about. It also meant I couldn’t break the server which happens when I start tinkering too much. I could rest easy knowing these folks had it all sorted.
When I had a bit more time, I moved to GitHub Pages to have a bit more control over the build. I really liked the idea of being able to make theme-related changes that take effect with a single
Not long after, I started to long for a bit more control over the functionality not only afforded by the front-end, but by what happens behind the curtain as well.
Headless CMS #
This is the new cool kid on the block, and I wanted to see what it was all about. After looking through a few options, I had a go of Contentful. I had a blog prototype set up pretty quickly with Vue.js as a single page application. It was pretty nifty if I do say so myself. I gave myself a pat on the back and thought "yep, this is it".
After about 20 minutes, however, I realised I needed server-side rendering to make it more accessible. I wanted to get the ball rolling quickly, and this was something that I felt would be a bit of a time sink. I scrapped it and decided to keep moving.
I’ve worked a lot with PHP-based projects including WordPress, ExpressionEngine and Laravel. I wanted to use something that I was familiar with in order to extend it with any custom functionality I might need/want. The Twig templating language is also something I’m quite fond of when it completely changed how I worked with WordPress. Lucky for me, this is where Craft came in.
I’ve used it before, and really liked how straight-forward it was to get going with. I thought about using WordPress, and while I still have great time for it, I wanted to explore something different.
Craft is also nearing it’s release of version 3, so I figured it was a good time to get back into it and see what’s changed since I last used it. Spoiler: it’s looking pretty frickin’ good. They also have a generous pricing plan for personal use: it’s free!
With that I pulled down the latest version of V3 that was available and started porting my posts over.
Laravel Mix #
Again, familiarity. Laravel Mix wraps itself around webpack like a blanket with a high thread count, and gives you a dead-simple way to get up and running quickly. If you need to dig deeper you can extend the configuration to suit your needs.
It was recently updated to include async/await straight out of the box, which is one less thing to hack about. I like not having to think sometimes, even if that saves me manually adding a dependency and updating a dotfile.
I stated earlier that I have a tendency to break things when I have the terminal open in front of me. Time progressed and since then I’ve learned enough to be dangerous. I signed back up with Digital Ocean because this time I had a secret weapon — Laravel Forge.
Forge is suitable for any PHP project, and provisions a server with a few clicks. It also lets me watch a branch on a repo and can deploy any time a push is detected. It’ll run a script as well, which means I can run things like
npm run production to create a production build of my assets.
Here We Are #
This has led to the site you’re currently on today. I can’t say it will always be this way, but for now I’m really happy with the end-result. You might notice the common theme of familiarity with these decisions. That was something that occurred after the ideas had settled. I wanted a solid base to work from, something that I had initial experience with to get something up and running quickly. Who’s to say I won’t end up with a headless CMS? Maybe when I learn more about server-side rendering I’ll pick up where I left off. Until then, I have more reasons than ever to keep tinkering and exploring what else I can do with this setup.