Which markup processor should I use? My needs for one happen to be simple: it should be under active development, should make embedding \(\LaTeX\) dead simple, and should be deployable with Markdown on Github Pages, since I plan to compose posts in prose.io. On offer, we have Redcarpet, RedCloth, Rdiscount, kramdown and Maruku.

• Maruku: The last commit was 4 years ago, I’ll give this a pass.
• Redcarpet: Word is that it’s wicked fast since it’s in C.
• RedCloth: This is at Textile processor.
• Rdiscount: Also written in C, supports a superset of Markdown.
• kramdown: Also supports a superset of Markdown, and plays well with embedded \(\LaTeX\) math!

I’m split between kramdown and Rdiscount. They’re both actively maintained and well documented, Let’s look at the interesting features they each provide:

• kramdown:
  1. Supports tables.
  2. Supports custom attributes for block and span elements (very useful).
  3. Supports footnotes.
  4. Supports ID’s for headers, useful for TOC’s linking to subheadings.
  5. Automatic TOC generation.
  6. Support block and span level embedded \(\LaTeX\), rendered via MathJax.
• Rdiscount:
  1. Supports paragraph centering.
  2. Supports image sizes.
  3. Supports custom attributes for block and span elements.
  4. Supports tables.

I’m giving kramdown a try, and am really liking its sensible syntax errors, native support for embedded math and custom attributes. I think I’ll stick with this for a while.