Contributing author: Deborah Pickett
Date: June 9, 2006
XML's a funny beast—people imagine that it has all kinds of mystical powers, but it's little more than a set of rules for how to structure data.
I'll try to work in a reference to the emperor's new clothes here.
If all you plan to do is write topics in DITA, then you need to know very little about XML. If you're going to use a newfangled WYSIWYG editor to hide
the structure from you, then you might be able to get by knowing nothing at all about XML. If you stay in plaintext-land, then I'd list the following XML ideas as essential:
- Elements and attributes
- Angle brackets (< and >), ampersand (&), slashes, and quotation marks (" and ")
- Content models (the idea that elements contain other elements)
These would be nice, too:
- Doctypes
- Well-formedness vs. validity
Don't let anyone try to talk you into studying these in the short term:
- DTDs and schemas
- XSLT
- Namespaces
After that, it's probably just a matter of brushing up on DITA's ideas of content models.
The above doesn't hold if you are planning to write your own specializations or transformations.
If you want to understand the internals of the DITA Open Toolkit, then you are in for a bumpier ride, because you'll need to become acquainted
with Ant, Java, XSLT, and how they all talk to each other.