Re: [DM] BEA_005

Michael,

First: our specs are inconsistent in tis area. Both the Data Model
and the XQuery specs are written as if the type annotation returns a
Qname. We HAVE to fix this problem.

Second, allowing the type annotation to be an anonymous identifier
seems like one of those  decisions that will cost us grief for the next 
10
versions of XQuery.

Even if in V1 this request for functionality will not be accepted:
http://lists.w3.org/Archives/Public/public-qt-comments/2004Feb/0426.html
it will be in some ulterior version. Our experience is that 
functionality is really
required for serious use cases.

And when it will be the time to accept this, then we will suffer the 
consequences of
this bad decision.

Best regards,
Dana



On Feb 16, 2004, at 2:39 AM, Michael Kay wrote:

>>
>> We should simplify the data model and request the name of a type (even
>> anonymous) to be a legal Qname.
>>
>
> Since there is no way for an application to determine the name of an
> anonymous type, this would be (a) an unnecessary constraint on
> implementations, and (b) an untestable assertion.
>
> Of course, if your comment requesting that this property be exposed to
> applications is accepted, then this objection disappears.
>
> Michael Kay
>

Received on Monday, 16 February 2004 14:55:46 UTC