Structured text or HTML

Posted March 28, 2006 in misc

Markup languages are based on annotating our content with some additional text to indicate the structure. Compared to plain text this is a certain amount of effort. And while you’re editing the document, your view of the content is also somewhat obscured by the “markup noise”. Could we not invent some more lightweight markup? One that deduces the structure from some subtle “cues” in the text itself?

This is the idea behind structured or humane text formats. 43Folders.com has a better explanation. I also recommend reading John Gruber’s advocacy article. He’s the author of Markdown (a popular structured text format).

Popularised by Ward’s Wiki and Perl’s POD format, you can read this comparison done by Thomas David Baker. Not everyone agrees on the conclusions though. This document by Mark D. Anderson is another good list of alternatives .

The weblog part of my website uses a combination of Ruby, YAML and Textile that I shamelessly stole from the great Hobix. It works for me, but I’m not sure I’d recommend it to everybody.

Tim Bray makes a good case that we should use one of the existing XML applications rather than invent our own. He also shows convincingly how XHTML+Microformats (using the div and span tags with the class attribute) can be used to produce XHTML that renders nicely (using CSS), but also has structural information at the same time.

If you’re trying to generate web pages, the structured text formats work nicely. But if you’re trying to generate XHTML+Microformats, I’m not so sure…