- From: Erik Wilde <dret@berkeley.edu>
- Date: Wed, 06 Feb 2013 16:54:45 +0100
- To: www-xml-xinclude-comments@w3.org
- CC: Jeni Tennison <jeni@jenitennison.com>, Michael Hausenblas <michael.hausenblas@gmail.com>
hello all.
the definition for fragid currently reads like this:
"fragid: The fragid attribute is a generalization of the xpointer
attribute. A fragid may be present regardless of the value of the parse
attribute. The interpretation of the value of the attribute depends on
the value of parse. For XML processing, the value is interpreted as an
XPointer (see [XPointer Framework]); for text processing, it is
interpreted as a [IETF RFC 5147] fragment identifier. For other values
of parse, the interpretation is implementation-defined."
maybe it should be a little less open in terms of allowing
implementations to do what they like. for example, we're in the process
of defining fragids for CSV
(http://tools.ietf.org/html/draft-hausenblas-csv-fragment). XInclude
implementers should look at all media types they support, and then check
for defined fragid schemes and support them, if there are any. if not,
maybe they should have the freedom to invent their own, but there
probably should be a requirement to implemented standardized fragid
semantics (if any are implemented).
cheers,
dret.
--
erik wilde | mailto:dret@berkeley.edu - tel:+1-510-2061079 |
| UC Berkeley - School of Information (ISchool) |
| http://dret.net/netdret http://twitter.com/dret |
Received on Wednesday, 6 February 2013 15:55:24 UTC