summaryrefslogtreecommitdiffstats
path: root/notes/markdown-vs-multimarkdown.txt
blob: 1d71d57640ba9af312b49238f9276bc5e4a76251 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Markdown: <http://daringfireball.net/projects/markdown/>
MultiMarkdown: <http://fletcherpenney.net/multimarkdown/>

MultiMarkdown seems to provide much of what we currently need and may want in
the future: cross references, footnotes, tables, and PDF rendering (though this
can be done with Markdown as well using Pandoc).  Why don't we just use that to
do all our work for us?

The cross references (hyperlinks to sections) only work within a single
document.  We want policy manuals to have a chapter/appendix per document (e.g.
HTML file).  MultiMarkdown doesn't really support hyperlinks to sections in
other documents.  So while MMD might be nice for its other features, it looks
like we should still handle in-book hyperlinks with our own pre-processing.

Then why don't we just use normal hyperlinks to sections in different chapters?
This assumes we're outputting documents only in HTML.  We would have source text
like this:

    See [Library Packages](binpkgs.html#librarypackages) for more information.

Rendering a plain text file that refers to other plain text files with a ".html"
extension makes no sense.  Additionally, this fails to work in the case of PDF
rendering.