This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
QT approved comment: Section 2.6.1.3 contains a mention of xsi:type, treating it as if it modified the mapping function from lexical values onto values of a union type. This would mean that the mapping from lexical space to value space is not actually a simple function as described in 2.4: it becomes a function that takes two parameters, the lexical value and the value of xsi:type. It would be better to treat xsi:type not as "overriding the evaluation order", but rather as causing a different datatype, and hence a different mapping function, to be invoked.
(In reply to comment #0) > QT approved comment: > > Section 2.6.1.3 contains a mention of xsi:type, treating it as if it > modified the mapping function from lexical values onto values of a union > type. This would mean that the mapping from lexical space to value space is > not actually a simple function as described in 2.4: We're correcting wording elsewhere so as to make the lexical mapping not necessarily a function see, e.g., bug 3227. Accordingly, I'm unilaterally marking this one wontfix and change declined. If the commenter does not agree with this resolution, please reopen the bug; if you do, please mark it closed. If there is no response within two weeks, we will assume the resolution is acceptable. (You can, of course, still object to the planned resolution of bug 3227.)