XML-Events Last Call Comments

Steven Pemberton, HTML WG Chair

(The numbers in brackets refers to the identification number in the tracking system)

Arnold Curt (623)

Eliminate the target attribute, and rename the observer attribute to 'target'.
It would be wrong to rename the observer attribute to 'target', because observer and target are two different concepts, and many things would break if you only listened on the target.

It would be possible to eliminate the target attribute (and retain the observer attribute), but there are cases where it is useful to specify the target as well.
No Qname
Agree. We have removed references to Qnames..
Discuss SVG events here.
We have added a mention at the appropriate place. Done.
Use the names as defined in the DOM spec
Agree. We now do not list them, but refer directly to the DOM2 Events spec for names and semantics. Done
List the DOM2 Events spec
Agree. Done
Use the same names as the DOM2 spec does
Agree. See above.
'handler' should be an idref (not a URI) and reference an <object>
No. 'Handler' should be able to reference an external handler. We don't define (in this spec) what it points to, to allow for maximum flexibility. A future companion spec will define options for handlers.
Include a schema
Agree. However, we are awaiting Modularization in Schemas to advance before doing this, so a schema will be added later.
(624) change mime type for ecmascript
http://lists.w3.org/Archives/Public/www-html-editor/2001OctDec/0446.html

We were planning on doing that, but a registration for ecmascript is not yet available.

There is a problem using ID/IDREF in the absence of a schema or DTD
This is true. There is nothing we can do about it however. It is an architectural problem of XML. We assume the User Agent can identify the elements.

"The implication is that if there is a time model, it must be SMIL's model, and default handler behaviour is that of SMIL's"
We don't agree that this is implied by the text. It is certainly not intended! The text has been modified accordingly.
Handlers have no predefined semantics. Language definitions should be required to define handler semantics.
Agree. Section 2.5 says this.
Event names quoted are non-WAI compliant. They should be more compliant.
It is not a task of this specification to define actual event names. We have removed all names from the specification, and referred to the DOM2 specification that defines them. We agree that this should be fixed, but not in XML Events.