I’ve had a vague sense for a while that I need to redesign this site, but procrastinated and avoided doing anything due to both time constraints and decision paralysis. (Also, 2020 was really not the year for coherently sustaining creative side-projects.)

What tipped me over the edge was sitting down with a bunch of new writing at the beginning of March 2021 and realising I was stuck. I had no clear pathway to publishing the work and did not know how to contain it. This wasn’t a problem with the writing of course, but with the website not supporting what I needed.

I’ve always like the philosophy of design in the open and have tried to practice it in various ways on lots of different projects in the past, but I’ve never applied it to the overall process of redesigning a unique content website with no commercial motivation. I discussed the idea of documenting the redesign process with a few people whose design opinions I trust and everyone encouraged me to run with it.

Given the current state of design discourse and the polluted nature of the web overall, it feels like my concerns about triviality and self-indulgence are outweighed by the value in sharing details of critical reflection and discovery on a real design process that addresses specific solutions to problems in context, as weird and unique as they might be.

Hopefully this documentation as a whole can also function as an argument reinforcing the value of focusing on the timeless fundamental materials and patterns of the medium, repudiating the technology hype and content marketing chaff that has swamped the practice of web design today and threatens to overwhelm it.

In the past, I’d start a redesign by forking the site, stripping out all the styling and templates, then get to work designing and converting content, working section by section, template by template. At the end of this process, I’d push it all live in one big bang (usually when it was unfinished but in a good enough state to park for a bit while I moved onto other things). I’d swoop in 2–3 months later with a big cleanup and fixup, and the site would settle into a pattern that would last for a few years.

My approach here is different. There won’t be a big bang or grand reveal. Every change I make will get released on the site as soon as practical. I’ll break down the ideation, analysis and decisions behind each step as I go, surfacing messy and discontinuous internal details—both creative and technical—that would usually be scrubbed away or hidden from view.

Cutting down the CSS

To avoid complications with assumptions around layout and to simplify auditing existing content and navigation, the first step in the redesign process is the visual reset. Cut down all the styles, leaving the basic HTML content and URL structure in place with browser defaults.

Because this is going onto the site as-is, I’ve gone a bit further in temporary affordances for legibility and readability beyond what I’d normally do in stripping out styling completely.

  • Set the body fonts and default size to 18px
  • Add breadcrumbs and basic link styling
  • Wrap the main content area with a max width of 960px (I’ll probably replace this with something adjusted for measure/line length, but I want to avoid getting caught up in typography and layout stuff right now, so I’m deliberately making a choice I don’t like as a reminder to stay focused on fixing the problems with structure)

I’m certain I’ve broken a bunch of essays and talks by doing this but that’s okay for now. I can always restore chunks of the legacy CSS for particular areas of the site as I go through. Obviously, this isn’t something I’d ever do with anyone else’s writing or on a client project, but as it’s only punching myself in the face, I am okay with that. Fixing things iteratively is going to save a lot of time.

As a requirements gathering excercise, it’s actually quite helpful to break things on purpose and see what falls out. These breakages quickly surface underlying markup details I’ve forgotten (it’s been a few years since I wrote these templates and styles) and reveal the scope of what I need to decide about archiving and putting containers and layouts in place for future resilience (I really don’t want to do this deeper planning work over and over in future, regardless of how frequently or infrequently I want to redesign layouts and typography).

With that done, the remaining thing before getting started is to rejig the homepage so that it explains what’s going on. Now that I’ve put it live, I’m committed, I will have to see where this goes.