This is the mail archive of the mailing list for the DocBook project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: [docbook] Amended DocBook Technical Committee Meeting Minutes: 20 April 2005

All well and fine that namespace != document type or schema location. After all, there is only so much you can assert with a single rather opaque symbol (such as an URI); I'd anyway often need a number of additional constraints to aid processing (such as validation).

It just strikes me as a bit silly that each and every consumer or producer of XML documents should invent and maintain each their own PIs/package format/catalog/whatever for basically the same purpose.

If the value of the version attribute is supposed to be helpful, surely we must have some protocol, some common expectations about how it is to be used? "lillet", "medium vodka dry martini, shaken, not stirred", "vesper", ""; and "7.00" will all validate; the version attribute is essentially even more opaque than a namespace. Is some sort of registry of version attribute values envisioned?

DocBook elements are going to be more welcome as guests in other document type families, now that they've got their own namespace. The version attribute is ubiquitous (belongs to db.common.attributes), and could be used to help a 'namespace router'; am I on the right track here?

BTW, vodka martinis taste like paint thinner if not ice cold.

Kind regards
Peter Ring

> -----Original Message-----
> From: Jirka Kosek []
> Sent: 26. april 2005 12:30
> To: Peter Ring
> Cc:
> Subject: Re: [docbook] Amended DocBook Technical Committee
> Meeting Minutes: 20 April 2005
> Peter Ring wrote:
> > So what would be the semantics of this DocBook namespace?
> > Given that "it would apply to the family of DocBook schemas"
> > (dare I read architectural form?), there's room for something
> > that asserts the applicability of a *specific* member of the
> > family to a document or parts of a document. Lots of application,
> > e.g. editors, are going to implement each their own mechanism
> > of asserting document types; some already do.
> Please note that document type and namespace are two
> different things.
> There will be several document types expressed as RELAX NG (and for
> compatibility with other tools also as DTD and W3C XML
> Schema) grammars.
> E.g. DocBook, Simplified DocBook, DocBook Slides, DocBook
> website. You
> will have four grammars, for schemas, but all elements will be in the
> same namespace.
> > I know about ongoing work in ISO/IEC 19757, but I
> > seem to miss the magic that will connect the pieces.
> Associating schema with document is thing which depends on tool used
> (for RELAX NG). So it depends on tool you will be using, but
> many tools
> allow you to place link to grammar inline as PI or global attribute.
> Some other tools use external configuration files to glue these two
> things together. But each DocBook derived document will be in DocBook
> namespace and will have version attribute. This should be enough to
> locate schema.
> --
> ------------------------------------------------------------------
>    Jirka Kosek     e-mail:
> ------------------------------------------------------------------
>    Profesionální školení a poradenství v oblasti technologií XML.
>       Podívejte se na náš nově spuštěný web
>         Podrobný přehled školení
> ------------------------------------------------------------------

To unsubscribe, e-mail:
For additional commands, e-mail:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]