During one of my dev classes, we had to do a simple web project. The topic was free, so I went with a blog engine. I could've knocked out something basic in three hours – database, some HTML and JS, done. But I figured: if I'm building one anyway, why not make it the one I actually needed?

The Old Blog Wasn't Really a Blog

The previous engine (still up at legacy.douxx.blog) didn't start as one. It was a documentation site, originally built for UrlToApp, a now-archived project. At some point I repurposed it into a blog, restyled it to match my main website, and started bolting things on – RSS feeds, comments, categories, the works.

The problem is that's exactly what it was: bolted on. Every new feature was stacked on top of code that was never meant to support it – cover images, for example, weren't proper metadata. It just scraped the article for a ![hero](<url>) tag. Not even an actual hero image. Eventually it got to the point where adding anything felt like defusing a bomb. Not fun.

So when the school project came up, the choice was obvious.

What Changed for Articles

A few things changed on the content side too. Some older articles just weren't worth keeping – they were outdated, half-baked, or both – so I cut them rather than migrate them for the sake of it.

The ones that made it over also got a fix that was long needed: inline JavaScript already existed in the old engine, but it was sluggish and held together with duct tape. It actually works properly now.

And if you followed an old link with ?p=something, don't worry – those redirect automatically to the legacy site, so nothing's broken.

End Note

I tried to make this transition as smooth as possible, but a few things didn't survive the architectural changes. Callout boxes – the little warning, tip, and note blocks with icons – are gone, and so are your old preferences if you had any set. Not huge losses, but worth mentioning.

If you want to set things up again, /settings is where you'd go.