A Year With Hugo

calendar Jan 1, 2026 · 3 min read
 en site  · leaving-wordpress
 ·
Share on: linkedin copy

Overview

Some thoughts on WP migration, but first of all, the Wordpress Drama status

Wordpress drama after an year

From a reddit comment around 5th October 2025:

[…] The October 2024 ACF plugin takeover was the canary in the coal mine. […] the actual migration to SaaS makes more sense than ever. Breaking free from WordPress means owning your stack, predictable subscription revenue, and not waking up to find your plugin forked because of a corporate pissing match.

The ecosystem isn’t dying, but trust is hemorrhaging. Developers who built businesses on WordPress are realizing the “open” part has terms and conditions written in invisible ink

As expected Wordpress openness is gone.

The darkly funny part? This accelerates the SaaS exodus Matt opposes. Smart money is already decoupling or optimizing to stay small enough to not matter.

Cited from another comment.

But now stop sadness: lets see what we have got from the 2025 switch!

Pros

  1. Automatic summary generation works great
  2. Isso Commenting server Rulez but you already know it. Also Isso is less prone to spam so far. This was one of the critical pain points.
  3. My attack surface is near to zero No more databases or PHP scripts running, no firewall, no strange security plugin for wordpress etc… a static site cannot be forced in any way.
  4. Ability to insert TeX math is provided also on WordPress, but it was easier to use in Hugo.
  5. For simple schematics, I find Decker a good tool: sharp, minimalistic and easy to integrate. Also it is very easy to generate html interactive decks. For other schemas and examples, GIMP is a very complete tool. I used it to make logos, images and so on.

Mixed

  1. I managed to get articles always in reverse-chronological order, and I also set up publish order on related-series list.
  2. Taxonomy configuration give me some troubles, because of my little knowledge of Hugo: in general it is very easy to create new taxonomies and I was able to build the “series” page. I needed time to figure out how to fix related pages, but it worked. Overall, GO templating language is manageable, even if sometimes I have trouble finding the answer in the Hugo documentation.

Cons

  1. Folder organization is a bit a mess when you have 23+ years of articles and you keep going for other 25 minimum. As usual a folder-content organization is less strong then a database one.

  2. Changing theme is more cumbersome than Wordpress, because it require far more amount of time. In this first installment, I ended up customizing a lot of pages, and theme replacement do not seems so easy to do.

New tools

To manage tags and categories of over 900 articles, I ended up creating a small python tool using click library and frontmatter library. The tool make the content very easy to edit in a massive way and it was easy to setup.

It replaces some wordpress plugin (like tag2category and category2tag converters) and compensate for the lack of a true database.

It required just half a day to setup and work very well.