Rebuilding my website
I decided to rebuild my website in Kirby CMS, reprocessing 1600+ lines of content across 20 years from micro.blog. Here I talk about what I have built, how the move went, and what I still have left to do.

It feels slightly crazy to have devoted most of April, and maybe a little of March, to a complete website rebuild, having only recently a few months ago spent the whole of my December leave building a custom theme for micro.blog. Why on Earth would I do that to myself?
Probably a number of reasons.
First, and mainly, while I made a small number of valuable connections at micro.blog, I just did not feel that I was getting much back. In an earlier post on my Scribbles newsletter I talked about how that conjured up memories of not being in the ‘in-crowd’ at school. I did not need that feeling again, and I’d actually have valued a little more interaction in what was a really tough year.
Second, as an advocate of POSSE - publishing on your own site syndicating elsewhere, I think that you really need to have control of your own technology stack. While micro.blog has some great apps and shortcuts, there are limits and compromises on the functionality within Hugo that you can use. To be fair, if I was enjoying the ride, and if I didn’t already have alternative hosting, I’d have probably lived with that, and my re theme in December had stretched the functionality about as much as I wanted to without being able to access the frontmatter. It wasn’t perfect, but I was happy with my build.
I decided at the end of March that I was going to rebuild my site, and after a short Processwire v Kirby CMS A/B test, I decided on the latter - mainly because I could see that the starterkit template already held most of the functionality that I wanted, whereas in Processwire I’d need to write my theme from scratch. If I’d realised how much time I’d spend on content vs theming, I might have gone the other way - especially given the pre-work that I’d put into fields and templates in a Processwire build.
To accelerate go live I decided to drop most of my workout posts from my site. In my new build I have constructed specific content types for routes, and activities on them. So I’ll start from scratch, and maybe rebuild data only from Strava. In future I’d like to be able to load gpx or fit files directly into the site and visualise here - though my technical skills are a long way away from being able to do that.







So a month later, and last night a got to the point where I could flick the switch and push the new site live. Though I have made alterations to font-face and weights, and I have created a couple of specific templates - the design remains very much Kirby CMS starterkit.
I like the format of the ‘blog’ page, and I’m planning to link ‘notes’ and ‘places’ together (when appropriate) in the same way that I am routes and activities. As I created specific content types too, it is only the ‘blog’ content that populates my feed.
I really like the magazine feel of the place pages - I still need to add much of the text into these pages, but I really like my focus on Boston’s stadium:
See the page here.
Anyway, enough talk about my site for now. I’m especially looking forward to adding new content to it rather than improving old content. For the first time in probably a decade, all of my images appear in the posts they are supposed to from the beginning of time. I hope that you like it, and if you do let me know using the kudos buttons below or on my Mastodon post circulated by EchoFeed.
Future plans: (!)
Backlog
-
implement a flavour of my ‘1987’ theme into this build
-
decide whether to implement one of the indieweb Kirby CMS plugins into the build or to continue using EchoFeed.
-
If I go with the latter, I will look to implement some privacy friendly comments system into the build.
-
I’ve deliberately left Tinylytics ‘Kudos’ function out for now. I may turn it back on at some point, but am trying to be mindful of recurring spend. Anyway, it’s still crunching analytics behind the scenes for now.
-
Work on closer integration / automation between content types as above.
-
I’ll probably remove hero images from the theme - let me know if you have strong opinions either way!
-
I almost certainly need to implement microformats into the theme too.
