Re: xml:rel, xml:href, xml:type

Paul Grosso <paul@paulgrosso.name> writes:
> Please consider commenting on my draft response below.
> I'd like to be able to send it right after this week's
> telcon at the latest.

Inline, below.

> On 2012-05-02 10:52, Paul Grosso wrote:
>> Here is my draft response to an xml-editor comment.
>>
>> I'm particularly unsure of my paragraph about associating
>> a MIME type with a link/resource, and I'd appreciate any
>> comments on my draft response.
>>
>> paul
>>
>>
>> -----------
>>
>> Eventually to: "Rushforth, Peter" <Peter.Rushforth@NRCan-RNCan.gc.ca>
>> Eventually cc: xml-editor@w3.org <xml-editor@w3.org>
>>
>> On 2012-04-06 19:48, Rushforth, Peter wrote:
>>> Hi,
>>>
>>> I think I previously posted this to the wrong list - apologies.
>>>
>>>
>>> Has it ever been considered to include standardized "typed links" 
>>> into the core xml specification, in a manner that is similar to 
>>> xml:base and xml:lang? (ie. via the xml: namespace).
>>>
>>> If these attributes were available to xml authors, xml could be a 
>>> hypertext language without schema designers having to reinvent the 
>>> markup for this concept; such re-invention  hinders interoperability.
>>>
>>> I propose the associated attributes below in the xml namespace
>>>
>>> xml:href - a single URI value
>>>
>>> xml:rel - one or more space-separated tokens or URIs, which identify 
>>> the relationships the targeted resource has to the current context.  
>>> The value of this element could be defined to be compatible with http 
>>> linkinghttp://tools.ietf.org/html/rfc5988
>>>
>>> xml:type - a single MIME media type, with a value compatible with
>>> http://www.ietf.org/rfc/rfc2046.txt?number=2046
>>>
>>>
>>> This would allow XML representations the ability to contain "typed 
>>> links", which is important for RESTful applications.
>>>
>>> If this idea is not stupid, how to go about proposing formally?
>>>
>>>
>>> Cheers,
>>> Peter Rushforth
>>
>>
>> Peter,
>>
>> Thank you for your interest in XML.
>>
>> The XML specification defines a mostly semantic-agnostic markup
>> technology that is relatively easy to parse and upon which many
>> higher level semantics can be built.
>>
>> Hypertext linking outside of the document is not something which
>> needs action at the parser level, so it doesn't need to be specified
>> in the core XML specification.
>>
>> The existing XLink Recommendation [1] already defines namespaced
>> markup for hypertext linking that would appear to address your
>> requirements.  In fact, XLink defines an xlink:href attribute
>> that appears to parallel your suggested xml:href attribute.
>>
>> Note that the xlink:type attribute is not parallel to your
>> suggested xml:type.  In fact, I don't see why a link should
>> have a MIME media type associated with it, since the URI in
>> the xlink:href attribute points to a resource, and the resource
>> determines the media type.

I think I might have said something like this:

  Note that the xlink:type attribute is not parallel to your suggested
  xml:type. Providing the MIME media type of the link target at the
  point of the link source is convenient for some applications but is
  inherently risky. The URI points to a resource, and retrieving a
  representation of that resource determines the media type.

but I think your response is fine.

>> Also note that the XLink specification allows the definition of
>> more complex linking relationships, though those extended features
>> need not be used for simpler links.
>>
>> paul
>>
>> Paul Grosso
>> for the XML Core WG
>>
>>
>> [1]http://www.w3.org/TR/xlink11/

Looks good. Sorry I didn't comment sooner.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh
Lead Engineer
MarkLogic Corporation
Phone: +1 413 624 6676
www.marklogic.com