This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
The specification of xml-to-json() states that whitespace text nodes are ignored. However, this doesn't make sense for text nodes that are children of a <j:string> element, since the text node child of j:string contains the value of the string, which may be whitespace. It should say "whitespace text nodes are ignored, except those whose parent element is an element named 'string'" Also, for the content of other simple-valued elements (number, boolean) I think it makes sense to strip leading and trailing whitespace, so that an untyped input is accepted if the result of validating it against the schema would be accepted.
The change was accepted by the WG on 10 Mar 2015 and has been applied to the spec.