OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   RE: [xml-dev] A thought on XLink in PIs for processing directives

[ Lists Home | Date Index | Thread Index ]

> From: Rob Lugt [mailto:roblugt@elcel.com]

> One problem with this idea is that PIs often have multiple pseudo
> parameters.  For example xml-stylesheet has the following 
> pseudo attributes
> defined:
> charset CDATA #IMPLIED
> alternate (yes|no) "no"

This *almost* fits into the XLink simple link model. Adding a "title"
pseudo-attribute, for instance, is consistent with XLink (since it allows
titles). Media or alternate could be addressed via an appropriate arcrole
(or "purpose" to use the RDDL term), though there is an obvious challenge,
here, if you want to specify both. Charset is sufficiently generalized that
one could specify that xml-link PIs can have an "encoding" attribute. So the
only thing that seems problematic, here, is how to specify 2 arcroles in the
instance you want to specify the target media and indicate that this is an
alternate. Is "alternate" sufficiently generalized to warrant its own
pseudo-attributes? If this concept of an "alternate" is relevant to other
types of resources, as well, then I think just elevate it to its own
pseudo-attribute, and then this works.

This seems to be a challenge for RDDL, though. In RDDL, the arcrole for a
link to an XSLT resource is the nature of the document produced by the
stylesheet. How would you specify media and/or alternate via RDDL in such a
scenario? If it's a CSS stylesheet, you could use the arcrole (presumably)
to specify media or alternate, but not both. Has anyone run up against this
as a problem while trying to use RDDL?


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS