W3C home > Mailing lists > Public > public-svg-wg@w3.org > October to December 2008


From: MURATA Makoto (FAMILY Given) <eb2m-mrt@asahi-net.or.jp>
Date: Thu, 23 Oct 2008 04:41:44 +0000
To: public-svg-wg@w3.org
Message-Id: <20081023133800.6CED.EB2M-MRT@asahi-net.or.jp>

Dear colleagues,

I would like to make some comments on the "DTD compatibility" constraint.

First, the spirit of RELAX NG is to separate infoset modification
from validation, which basically reports either "valid" or "invalid".
Assigning IDness to attributes is infoset modification.  Thus, the
RELAX NG specification does not cover IDness.  Many programs (e.g.,
ODF editors) require IDness, although they do not perform validation.

The RELAX NG DTD Compatibility specification is intended to help
migration from the DTD language to RELAX NG.  However, this spec is
not an OASIS standard or ISO/IEC standard, and has not been fully
implemented (jing does not implement default values, although it does
implement IDness).

I thus think that the best approach is to rely on the xml:id
recommendation WITHOUT USING RELAX NG VALIDATION.  The xml:id processor
provides IDness.  I would recommend s/ID.datatype/NCName.datatype/g.

You might be interested in a recent e-mail thread about this topic,
available at:


Hope this helps.


MURATA Makoto (FAMILY Given) <EB2M-MRT@asahi-net.or.jp>
Received on Thursday, 23 October 2008 07:21:01 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:29:40 UTC