- From: Jirka Kosek <jirka@kosek.cz>
- Date: Tue, 13 Nov 2012 22:51:28 +0100
- To: public-xml-core-wg@w3.org
- Message-ID: <50A2C0E0.3080908@kosek.cz>
Hi, we had discussion about reserved names starting with [Xx][Mm][Ll] during TPAC face-to-face. Main reason was that in recent time I have came across several usages of such named elements. Purpose of this email to get feedback about issue from people not attending face-to-face. Main points of discussion are captured in minutes available at http://lists.w3.org/Archives/Public/public-xml-core-wg/2012Oct/0059.html There are several issues related to such names: 1. Wrong statement about source of reserved names in XML Rec ============================================================ XML Rec says that: "Names beginning with the string "xml", or with any string which would match (('X'|'x') ('M'|'m') ('L'|'l')), are reserved for standardization in this or future versions of this specification." However examples like xml-stylesheet, xml-model or even XML namespaces (xmlns:...) prove that reserved names are usually defined in separate specification. On F2F there was consensus to adjust wording and say "... in this or future specifications from the XML Core WG or its successors" If there is agreement about such change, we yet have to find best way how to implement this change (erratum, new edition, ...). 2. No guidance how to deal with names that are starting with xml ================================================================ In past same XML parsers emitted warning when they came across element/attribute/PI which was starting with [Xx][Mm][Ll]. However majority of tools simply doesn't care. This is problem especially in applications like XML schema editors -- they are not indicating that such name is wrong. This might be caused by the fact that XML Rec is not giving enough guidenance in this regard. Should spec explicitly say that usage of reserved name has to be reported? Should all application report this only those who are producing XML or XML schemas? Is that practical as list of reserved names can evolve over the time and not all software component are updated regularly?s 3. Are local names starting with [Xx][Mm][Ll] allowed ===================================================== We haven't touched this during F2F, but it is related issues. For example the following two formats are using reserved names for local names: http://www.loc.gov/standards/mets/docs/mets.v1-9.html#xmlData http://blogs.msdn.com/b/ericwhite/archive/2008/09/29/the-flat-opc-format.aspx We got similar inquiry year and half ago, see thread: http://lists.w3.org/Archives/Public/public-xml-core-wg/2011May/0007.html I think that there was no single and clear conclusion whether xml is reserved when used as a start of local name on element with prefix, like: <foo:xmlData xmlns:foo="bar"/> It's probably clear that default namespaces can't be used in this case: <xmlData xmlns="bar"/> Some clarification or guidence could be added into XML Names Rec. 4. Should we release reservation of names starting with [Xx][Mm][Ll] ==================================================================== There are documents out which are using reserved names already. We can choose two ways how to deal with this: a) Such names were always reserved, it's not our problem, it's your fault. We might only give more guidenance as outlined above. b) Since we have namespaces there is no further need to reserve names starting with [Xx][Mm][Ll]. We can use prefix xml: for future reserved elements and attributes and prefix xml- for future processing instructions. We will allow names starting with [Xx][Mm][Ll] with the exception of xml: and xml-. Advantage of this approach is that we would decrease divergence between XML Rec and reality. Any thoughts welcomed. Jirka -- ------------------------------------------------------------------ Jirka Kosek e-mail: jirka@kosek.cz http://xmlguru.cz ------------------------------------------------------------------ Professional XML consulting and training services DocBook customization, custom XSLT/XSL-FO document processing ------------------------------------------------------------------ OASIS DocBook TC member, W3C Invited Expert, ISO JTC1/SC34 member ------------------------------------------------------------------
Received on Tuesday, 13 November 2012 21:51:57 UTC