[ivanhoe]

Never really understood why people try to avoid DBs? It's OK for a very basic pile-of-online-texts blog, but the moment you try to show whatever relation between blogposts (e.g. the related articles or the latest articles) you end up reinventing the wheel

[vidarh]

Most blogs are small enough that caching all the content in memory and sorting/selecting is so cheap and simple it really makes no difference these days, and you gain simplicity.

My blog is flat file because I like to work on a version on my home server, editing stuff in emacs, commit to git and push an updated version atomically. Even if I continue writing at my current pace for the next 100 years, my current server would hardly notice having to re-read every single article.

As for reinventing the wheel, the code for pulling in the articles from flat files and slicing and dicing them simply by iterating over an in memory connection is so small and simple that there's hardly any wheel to re-invent.