RE: document node attributes

Hi David,

> 1) A new media type with metadata which is not part of the document body (outside the root)
> --> As Liam mentioned this won't survive XPath/XSLT/XQuery/XProc etc.

Yes I agree that one needs to have those technologies' support.  However, if a "NeoXML" parser has a namespace "built-in", then as a follow up question, would virtual properties on the document node be available to applications (XPath et al)?

> 2) Using something like automatic namespaces
> --> IMHO the cart before the horse.  We first need to get *that* spec adopted and commonly used, then it might > be great.

I see it as in the correct order.  Since the automatic namespace proposal needs a linking attribute added to the xml: namespace (xml:ns), why not add general links to the xml: namespace which then enables the automatic namespace functionality.  I would hope there would be many many such applications of links which have the proper 'level' or layered relationship then. All due respect to Liam's automatic namespace idea of course, which is in itself excellent.

> 3) Discussion why namespaces are bad  AND > 4) Discussion of micro xml.
> --> Micro xml ... got rid of *all* namespaces.  If we go that route then a Micro XML add-on
might be useful.   But it's not XML.

I did mention on the MicroXML CG list that eliminating the xml: namespace was a mistake, and I stand by that.
But I very much doubt that anyone in the MicroXML CG (besides me) is going to suggest that @href, @src etc be declared by the media type to mean anything, which means they won't mean anything.  At least they had a chance to be tied to link semantics via the xml: namespace option.  So like XML, we'll have a markup media type on the web with no semantics (even less, actually) apart from angle brackets and attributes.  I guess that's ok too, but it's not too useful above POX, IMHO.

Cheers,
Peter

Received on Saturday, 15 June 2013 00:54:14 UTC