About the future format for LFS (reworded)

William Immendorf computerperson1 at live.com
Sun Sep 7 15:59:41 PDT 2008

Matthew Burgess wrote:

>> DocBook V5.0 offers new functionality. DocBook V5.0 provides significant
>> improvements over DocBook V4.x. For example there is general markup for
>> annotations, a new and flexible system for linking, and unified markup for
>> information sections using the info element.

>Agreed, and this is why we normally upgrade to the latest version of DocBook when time permits.  And this is why I wouldn't have any problems in committing patches on someone's>behalf that upgrade the book to DocBook-XML 5.0.

Yup, that is a reason.

>> DocBook V5.0 is easier to customize. RELAX NG offers many powerful
>> constructs that make customization much easier than it would be using a DTD
>> (see the section called “Customizing DocBook V5.0”)."

>But the crucial point you're not making clear is *why* does LFS require that customization feature?  What is it you think LFS will gain by such customization?  And why can that not be>gained without the customization feature of RNG?

Why? Here are 2 reasons:

1. We can create new elements to replase repetive elements. An example of this is the para-screen-userinput combo we use for commands. Like, we can replase that with a code element with a description element in it.

2. We can also intergate elements from other XML languages like SVG, MathML, XHTML, or even *ML.

Also, here are 3 reasons we shouldn't use DTDs:

1. It has no support for newer features of XML, including namespaces.

2. It lacks expressiveness.

3. It uses a custom non-XML syntax, inherited from SGML, to describe the schema.

The only choices left are Relax NG and XSD, and I chose Relax NG because it is better than XSD.

For validation, we should also use Schematron for validation. The valdition language we will use is NVDL.
Want to do more with Windows Live? Learn “10 hidden secrets” from Jamie.

More information about the lfs-dev mailing list