The W3C released a Last Call Working Draft of Canonical XML. The draft includes a request that developers address a controversy over the use of the W3C Character Model for the World Wide Web.

The W3C's Character Model, itself a working draft, provides a set of rules and understandings for processing Unicode in the context of the Web. According to John Cowan's reply, the XML Core WG made this decision 'by vote, but not by consensus', but claims that 'overhead of normalization is not large in code space or data space or time'. (He also provides a more technical description of what is involved in another message.)

James Clark's minority report argues that this move both adds complexity and 'has a significant chance of splitting the market.' Given that digital signatures work is likely to rest on canonicalization, such a split could introduce lots of new problems, if it occurred.

Comments on this draft should be sent to www-xml-canonicalization-comments@w3.org .