At 02:00 -0400 8/6/07, Diane Gaskill wrote: >Now I have another question, almost as complicated. Structured FM can work >with both OASIS XML standards: DocBook and DITA. But which one is the better >standard to use? Or does it make any difference? One of our divisions in >Japan has decided on AT with Docbook 4.2 because it is what one of our OEM >customers use. They want us to do the same. We would rather use FM, but are >undecided on which standard to use.
Diana... I am sure you will get some responses to this that are far more authoritative than mine. However, to kick things off, it's probably worth pointing out that DITA and DocBook are not really directly comparable: DocBook is a general-purpose schema for structured technical documentation, but DITA is a whole documentation methodology that starts at the level of information chunking, or 'topics' in DITA-speak. The 'architecture' in DITA is the key term. Here are a couple of relevant references: <http://www.docbook.org/whatis> <http://www.ibm.com/developerworks/xml/library/x-dita1/> Both are 'standards' in that they are maintained and documented by a body appointed for that purpose, but both will almost certainly require adaptation ('specialization' is the term used in DITA) to make them suitable for your specific application. If you decide to follow the DocBook route, you might be interested in the packaged DocBook product, DocFrame, from Scriptorium Press, which is designed to take some of the pain out of using DocBook with FrameMaker: <http://www.scriptorium.com/docframe/index.html> I have no commercial connection with Scriptorium, although I have used their products (but not DocFrame). Others here may have direct experience. There is, as always, A Third Way, which is to create your own dedicated EDD. Deciding between these options is not easy, and is going to be highly dependent on your application, your need for 'standardisation' and compatibility. Do bear in mind, though, that the DocBook or DITA schemas are only intended as a fixed point of departure,which sort of colors the 'standardization' argument. -- Steve