DOCBOOK: Re: [docbook-tc] Proposal: Linking in DocBook

[Norman Walsh <ndw at nwalsh dot com>]

/ Peter Eisentraut <peter_e@gmx.net> was heard to say:
| Norman Walsh writes:
|
|> Fork 2: Do It Our Way OR Do It the XLink Way
|>
|>   If we do it our way, we get to define the semantics, but no tool will
|>   ever support our linking elements directly. (Well, I suppose some
|>   special purpose DocBook tool might, but let's not worry about that.)
|
| I'm wondering how such "direct" support would look like.  All toolsets
| that work with DocBook are in some way specially purposed for DocBook.

That's just not true. I have tools that work with any well-formed XML
document. Some of them do fairly trivial things, like count words and
scramble #PCDATA, but they don't care what markup is used. The
advantage of XLink here is that I could write similar tools for doing
link analysis or harvesting and they would work with any vocabulary
that used XLink for linking semantics.

| Today we have XSLT and DSSSL stylesheets (and other obscure conversion
| tools) that work specially with DocBook.  Even in the unforeseeable future
| there will have to be some sort of tool that associates semantics to raw
| DocBook.

For presentation, yes, but not for linking if the world adopts
standardized linking markup. Similarly, if the world adopts SVG and/or
MathML, you could write schema-agnostic tools to extract drawings or
equations from XML documents.

| For such a tool it's pretty irrelevant whether it converts xlink
| or some other linking system for presentation.

True.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com>      | Abandoning rhyme and fixed rules
http://www.oasis-open.org/docbook/ | in favor of other intuitive rules
Chair, DocBook Technical Committee | brings us back to fixed rules and
                                   | to rhyme with renewed
                                   | respect.--Jean Cocteau