So as you've probably noticed, this site is quite spartan. The reasoning and logic behind that decision is rather straightforward:

• I wanted to implement a .NET razor page(s) based site in my favorite programming language (C#).
• I believe SQLite is an excellent DB engine that can scale quite well for a simple site such as this one.
• I really desired something that was drop-dead simple that allowed me to authenticate and authorize users (authn and authz), and had an interface to post/edit/manage/delete categories, posts, and featured posts quickly.
• A curiousity for playing with ChatGPT, and seeing how much it could actually help (total implementation time was about 10 hours, ChatGPT probably saved me 10, but cost me 3 as well).
• A strong preference for keeping things simple (i.e. no ads, clean text, minimal design).
• I am not a front-end person, and this was the best I could do on short notice, without a graphic designer, with a design that was built from scratch to be functional without getting in the way.

Will I improve the site's underlying code and the design? Probably. I always seem to have an itch to scratch. I'm sure some things will become annoying or will not be as functional as I imagined, so there will be tweaks here and there. Already since penning this article, I have added MathJax support, and it occurred to me that I might have wanted an active/archived flag for posts, so I went ahead and implemented that too.

The site was written in modern C#, targets the Asp.NET Core 9.0 runtime, and runs on Ubuntu 24.04 with a SQLite backend. The SQLite database is used via Entity Framework Core (efcore) and uses the code-first method of DB generation (so the database and its migrations are built by efcore itself straight from the code via cli). The site is very fast and quite responsive.

Truth be told, this setup will probably handle all of the traffic I could ever hope to throw at it. It supports categorization of posts, featured posts, and allows me to do my editing in a WYSIWYG editor (TinyMCE). Getting code blocks working was probably the biggest roadblock and I went through 3 editors (Quill, TipTap, and finally TinyMCE LGPL). That whole escapade alone ate up around 4 full hours of implementation time.

Is there anything I would add? I thought about post tags, and adding a search feature based on them, but it seems superfluous when I can just create categories, especially since the categories jump right to the relevant posts. Overkill, essentially. I also contemplated comments, but that requires a bit more work, and would require moderation on my part. For the time being those features are on the back burner, because I've got to get some content with real meat & potatoes behind it first, and because anyone who needs / wants to comment can reach me directly via email.

I'd be happy to share the site's code or to discuss enhancements if anyone has suggestions. The code is not on GitHub because it is not ready for public distribution due to a few hard-coded values and the database seeding code.

Please enjoy!

-padawan, 2025