I'm dividing this post into two parts: the prosaic "how" and the maybe-more-interesting "why".
This blog is delivered to you by the Cloudflare Content
Delivery Network (or CDN) via their free plan, which enables
SSL, asset minification, and 1-2 page rules. The page rules
http://neversaw.us ("hot potato") redirects to
https://neversaw.us ("hot potatoes"), and and
https://neversaw.us redirects to
Of course, I'm not writing HTML right now; we'll cover how what I'm typing in my editor gets to you shortly though. First, let's review how I set these things up!
I provision all of my infrastructure resources with Terraform, and you can see how it's configured here. My personal infrastructure repo uses direnv to automatically install all of the tools I need. You can see that configuration here. I use 1Password to store my credentials, and load them into my shell on demand by sourcing this file. The file uses the 1Password CLI to access a well-known key in my vault, containing my AWS credentials in the form of Bash environment variables.
Now, as for my editing interface: I'm writing this in a folder on my mac laptop
using NeoVim. The folder is actually symlinked from
~/Dropbox/Apps/neversawus-blog. Whenever I edit files in this blog directory
the development Dropbox application webhook sends a request to a URL I've
configured, backed by an AWS API Gateway endpoint. This calls an AWS Lambda
that sends GitHub repository dispatch events to my
(The Lambda and Gateway setup are in my Terraform configuration, but they're pretty much taken from this excellent Hashicorp guide.)
chrisdickinson/neversaw.us receives a dispatch event, it runs a GitHub
Action. The action generates this site by downloading the current state from
Dropbox, then running it through Zola, a static site generator written in Rust.
(The Rust part isn't actually important except that I really enjoy Rust and want more
people to try it out. It's part of the reason I selected Zola, but I'm happy to
report I'm very happy with all of Zola's features thus far!)
There's a potential for race conditions, but thus far I haven't really run into any -- and in any case, I'm one "Save file" action away from fixing the problem.
And that's how you're seeing what you're seeing right now.
Programmers rebuild their blogs often. I, personally, have rebuilt this blog about five or six times. Each prior time, I focused on the technical features I wanted and lost sight of the experience & content I wanted to create. This is nearly a trope for programmers.
When building a documentation generator for Boltzmann, I found that Zola had all the technical features I had always wanted, which got me thinking: I had always (re)built my blog from the perspective of these features, and I rarely blogged in it after the fact. The writing is the point of the blog, and for whatever reason I had always stopped short.
A few things were floating around my head at this point.
I had lunch with a friend a few months back. Over a gigantic breakfast burrito, he mentioned that he was moving off of Twitter because it was taking up too much of his writing attention.
Later, I saw this tweet:
git is just dropbox where you have to write a tweet before it syncs, don't judge me
And it clicked for me. I wasn't writing in my blog because I was focusing my attention elsewhere, namely Twitter and Slack. Twitter's innovation was in reducing the friction between having a thought and communicating that thought to the world, no matter how ill-advised it might be.
On the other hand, writing a blog post was a ceremony for me. I'd write it, edit it, make sure the CSS and JS were perfect, then I'd commit it to Git and push it up. In the process I'd get so distracted by the framing of the content that I'd usually forget to keep writing content.
It has to be easier for me to write in my blog than to write a tweet.
Which leads us to here -- I write in a text file, save it, it gets rendered and uploaded, and now you're reading it. The framing is not great, BUT. The other brainwave I had was this: I kept redesigning my blog before I had any content in it. I was trying to create a design system ex nihilo; no wonder it had never felt right! I am forcing myself to feel discomfort about it until I have enough content to warrant framing it differently.
In Gall's words:
A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works and cannot be patched up to make it work. You have to start over with a working simple system.
I am starting a simple system of creating content. I will improve the styling and framing of that content slowly as I build more content. All of my prior attempts were focused on features and styling for content that wasn't there yet. I'm only four days into this experiment, but I'm (foolishly) optimistic.
Being able to type in my favorite editor like this and reach you on a platform I control -- that I'm not renting from Twitter, where I'm not taking space from other people who need it more -- that feels like a sea change to me.