This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
Some specifications have been posting DTDs in various locations. e.g: http://www.w3.org/TR/MathML2/dtd/xhtml-math11-f.dtd http://www.w3.org/Math/DTD/mathml2/xhtml-math11-f.dtd The validator currently has a single acceptable System Identifier in store for each public identifier, and will throw a warning if the SI does not match the one in store. We should change that to allow any of the acceptable SIs to pass.
What is holding up this bug?
This would be easy enough to do, but I'm unable to find a MathML specification that would somehow recommend the former of the two URLs.
As noted in: Changes to Appendix A Parsing MathML. of section: J.1 Changes between MathML 2.0 and MathML 2.0 Second Edition of Mathematical Markup Language (MathML) Version 2.0 (Second Edition) see: http://www.w3.org/TR/MathML2/appendixj.html#changes.mathml2.0-2.02e "The location of the MathML DTDs described in Section A.2.1 DOCTYPE declaration for MathML and Section A.2.3 MathML as a DTD Module are now in the Math Working group area rather than distributed with this Recommendation." The old recommendation was: http://www.w3.org/TR/MathML2/dtd/xhtml-math11-f.dtd the news recommendation is: http://www.w3.org/Math/DTD/mathml2/xhtml-math11-f.dtd The question is, should a document be validated if it follows the old recommendation, or should it be rejected unless it follows the latest recommendation?
Thank you for the reference! (In reply to comment #3) > The question is, should a document be validated if it follows the old > recommendation, or should it be rejected unless it follows the latest > recommendation? The way I see it, what the system id is does not matter wrt. validity. This is how it currently is in the validator; validator does not "reject" documents that use the old system id, it just issues a warning about it. I think this is just fine, and I'm inclined to leave things as they are and close this bug. I don't see anything in the MathML specification that would actually *mandate* using any particular system id - it's a "should" in a section that is not marked as normative unless I missed something. However, the warning emitted by validator points to a document that has the old location for the MathML DTD: http://www.w3.org/QA/2002/04/valid-dtd-list.html I'll try to find someone who can update it.