Re: Error Handling

Tim Bray wrote:
> 
> At 01:11 AM 8/16/98 -0700, David Brownell wrote:
> >Error handling isn't well enough specified in the current XML namespaces
> >draft.  If this is truly expected to be incorporated into an upcoming XML
> >revision (1.1 or so) this must be defined using the same terminology
> >(warning, error, fatal error).
> 
> I don't agree with David's conclusion, but he has characterized the
> symptom accurately. I have just gone through and reworded all discussion
> of errors simply to talk about things either being "conformant with
> this specification" or not.  -Tim

The conclusion only followed from an assumption -- is the assumption
still intended to be valid?

If the assumption is valid, then the conformance requirement will need
to apply to the XML processor, and implementors would need to know what
category of error to report.

Else, it's a document requirement that one could realistically expect
XML processors to ignore.  Which would be much goodness.

- Dave