thecashewtrader's Website

jump to main content

In Appreciation Of Weblorg

1 Introduction

I just set this blog up with weblorg and it has been a pleasant experience. This will be a short post explaining the differences I noticed from other Static Site Generators and reasons for choosing weblorg.

2 The Good

The Stack
Weblorg uses pure Emacs Lisp, and advertises this on it's website. This was more helpful than I expected though, because it makes stuff like CI/CD much simpler.
Weblorg doesn't make assumptions about project structure and has a very concise syntax for describing your (custom) structure.
RSS was something I required in a Static Site Generator, and Weblorg delivered. I was initially skeptical of builtin RSS functionality, but so far I haven't run into any issues.

3 The Bad

There's not really much wrong with weblorg, but I'll list some minor problems I ran into:

Lack Of Documentation
The documentation doesn't covert global variables which are used to populate the title, description, and keywords of the site are not mentioned in the documentation.
Default Template Incomplete
Following the default build script and RSS feed templates leads to empty or "default" (eg. "Simple Blog") values for these. This ties in to documentation about site global variables, and both of these are trivial to fix.

4 My Usage

I have 3 main org mode repositories:

|-- blog
|-- braindump
`-- roam
This is a private "master" repo containing my Org Roam files.
This is a repo that has public files from the roam repo.
This is a repo that has blog entries from the roam repo.

I have two scripts in my scripts repo to handle this through filetags.