- From: Elliotte Harold <elharo@metalab.unc.edu>
- Date: Wed, 04 May 2005 12:10:52 -0400
- To: Webb Roberts <webb.roberts@gtri.gatech.edu>
- CC: public-xml-id@w3.org
Webb Roberts wrote: > They assumed (correctly) that other attributes would be added to the > namespace, but were wrong in believing that they would support copying. > An errata could be added to the canonicalization spec that clarified the > point, and make explicit exactly which attributes from the xml namespace > are merged in. This is not a simple mistake that could be fixed by an erratum. It was a deliberate design decision, almost certainly the wrong decision but a decision nonetheless. It is irresponsible and dangerous to change the existing spec and suddenly invalidate tools that correctly implement the specification as published. Not that the W3C hasn't done this in the past, but it's been a royal mess when they have. We can't rewrite history. Possibly your use case could be adequately addressed simply by requiring the use of exclusive XML canonicalization. The existing algorithm really can't be changed at this point. New algorithms can be created, but that seems unnecessary when exclusive canonicalization already solves the problem. Is there some reason you can't mandate exclusive XML canonicalization? -- Elliotte Rusty Harold elharo@metalab.unc.edu XML in a Nutshell 3rd Edition Just Published! http://www.cafeconleche.org/books/xian3/ http://www.amazon.com/exec/obidos/ISBN=0596007647/cafeaulaitA/ref=nosim
Received on Wednesday, 4 May 2005 16:11:03 UTC