wykwit.pl / blog / blog-evolution



Blog evolution

2025-11-02


For years I've been rocking the same minimal design on my blog. I've been creating web content of some sort since I first learnt how to make basic websites back in primary school. I feel like the form of my personal website became inadequate with passing time and as my content continues evolving. It was time for a little make over. Most importantly, I had to improve meta info for my posts and unify the style of all pages. I've decided to focus on functional improvements rather than bold visual changes.

My little blogging evolution

When I first started this blog I had a different goal in mind than today. I wanted to make my dense notes public on the web so that I could refer to them whenever I needed. Let's take "Static IP" blog post as an example. There is almost no introduction, the post starts with a link to relevant Arch Wiki article, then there's a few command examples with short descriptions, and that's it. If I ever forget how to use the 'ip' command (which regularly happens) I can simply go back to that very post and the list of all the commands I need is right there. My blog was a collection of notes. Titles were as short as possible, I had no need for subtitles or catchy headers, only enough information for me to find what I knew I had saved. What little design there was served that purpose very well. However, it was not very approachable for others.

Before 2018 my blog was more of a little personal journal and that was... cringy. Some things are better left off the internet. Now I write my personal notes on paper, those can be ephemeral. I publish more deliberately, being concious about the fact that things here may stay alive longer and are generally available to see for everyone with no restrictions. I always try to keep some hard technical info in my posts. I still use the blog to note down things I would like to refer to in the future, but my old style of strict notes does not work with the things I learn anymore. I want to share more things I find particularly interesting and use this platform for more than just stashing notes. With the shift to more opinionated articles the form does not serve the function well enough anymore. I once again find the need for subsections, descriptions, and meta info for each page on the blog.

I published my blog on neocities over 5 years ago. Neocities websites were always my inspiration, and especially for my "about me" page back then. I really liked the way it turned out - minimal, clean and informative. However, it did not fit in stylistically with the rest. Regardless, I liked it so much that it just had to stay.

One of the redesign goals was to unify the style across all pages on this website: blog (article list), post (article content), network (links list), and the "about me" page. I went with the minimalist dark look from my blog, and adapted the "about me" page to better fit in with that style.

Design principles

Core design principles did not change. I need my website to be readable without JS, accessible from a terminal browser, very hacker-friendly. I want to use a dark theme by default. There should be no unnecessary distractions on the page. Content is the focus. If there is JS, it's only there to help the user, not as a requirement to view the website.

Frontend was never my strongest suite, but recently I came around to using modern hybrid web frameworks and I completely fell in love with Nuxt, which is built on top of Vue.js. I've been debating whether to use Nuxt or a different SSG for this website. I'm sure I want to use Nuxt for other websites in the future, but for this one a simpler SSG like Zola seemed to be a better fit. Until now I've been using JS to fetch the post content on the blog page, with links to text-only versions of the post in case the user does not support JS. I would re-generate JSON with a list of all posts on every publish with a little python script in my website repo. I could just as well use some SSG by hand or regenerate the website in a continous delivery pipeline on GitLab when I push to master. Maintainability should not really be a concern with statically generated pages.

I considered including some Visual Sugar: animations, page fade-in, code syntax highlighting. There are no long animations, so users don't have to wait for the content to display for too many millisecons. Adding syntax higlighting for code snippets was long overdue.

I was also debating whether or not to include more links on the home page, to things such as: recent posts, recently updated projects, maybe fediverse activity. Right now the home page is a simple landing page hub from which you can go to my blog or the "about me" page, and only from there do you get to posts and other places. The landing page gets many accidental visits and bot traffic, so I prefer to have the blog at least one click away from the home page. On one hand I like how minimal it is, but on the oter hand - it makes sense to better display what I'm doing here. As of now I haven't made that move yet.

Finally - metrics. I changed my mind about this, because for years I was strongly against it. I'd never have any tracking on the website, not even to count the page view. This time I would prefer to know which pages and posts attract most eyeballs and how many people are even reading this. For this I decided to use Umami analytics - it's very simple, self-hosted, open-source, privacy-respecting, no-cookie-bullshit, presumably GDPR and PECR compliant, easy to block if you care, and at the same time should give me enough insight for all I need.

Eventually I might also include a comments section for each article. If I built out a little dynamic service to gather user feedback, it would bring even more valuable insight, and make the website that much more alive. Still, I don't want to make any promises and I need to think it through a little more.

Results

When I first approached the idea of a redesign my imagination was running wild with all the things I could do. It was on my mind for over two months. The scope was so broad that no work was ever going to get done.

Only when I embraced small, focused and purposeful changes, did I make any progress in this effort. Instead of re-building from nothing, I took what I already had and improved it.

For now I'm pretty happy with how this website presents itself online. Migrating to Zola, a proper SSG, was a chance to fix many of it's previous technical shortcomings. These improvements compound, even when the effect is not a drastic unrecognizable change.