[
Lists Home |
Date Index |
Thread Index
]
On Mon, Mar 04, 2002 at 01:13:22AM -0800, Tim Bray wrote:
> Daniel Veillard wrote:
>
> >>Despite the fact that XInclude seems like an obviously-good idea, I
> >>do not observe a rush of implementations nor people complaining about
> >>the recommendation not being done. I'm not sure, to be honest, if
> >>the world really eneds it. I don't for any of my work. -Tim
> >
> > I have at least one large user relying on it. They do DocBook
> > processing and XInclude allows to split the document like external
> > parsed entities. But the XInclude chunks can be DTD validated
> > directly contrary to entities.
>
>
> Could you achieve the same effect with a perl script?
yes you can also validate and probably 100x time faster a document
directly by accessing the node tree with a high level language. And
I will argue an awful lot of people still do this too.
> I'm beginning to think that trying to do cheap content management
> down in the markup just doesn't have a positive payoff. -Tim
It's just a matter of API level. As parts get used more they
gets forged as API. Better building API at a syntax level than
in the programing language syntax. I call this sedimentation...
Daniel
--
Daniel Veillard | Red Hat Network https://rhn.redhat.com/
veillard@redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/
|
