Re: [xml-dev] XLINK support in browsers

[Razvan MIHAIU <mihaiu@mihaiu.name>]

Michael Kay wrote:

    XLINK is primarily about defining links between documents in a standard manner and not about defining relations  

 

Can you explain the difference between a link and a relation[ship]?

    In my understanding, in the XML world a relation[ship] is a link with meta information, like a parent-child relationship. When you define a link in XLINK you are not forced to say anything about the relationship between the linked elements. You state that there is a link between certain elements but you may choose not to enter into details - e.g. not to define their relationship. This is in my understanding  the difference between the 2 concepts: I hope that this isn't a non-sense to you.

    My understanding on the 'relationship' in the XML Core specification is slightly different - but that is the subject of another discussion.

 

    Honestly I do not understand your complain here: XLINK is a specification, so, like any standard it needs some fixed attribute or element name. The same goes for all the specifications. Lets take the XML Schema specification: you have some standard tags like "complexType, simpleType, complexContent" a.s.o. You could complain that XML Schema is useless because it forces you to use a certain set of tags.

 

If XLink is in the user interface space then it's reasonable for it to have its own vocabulary. But then it needs to be part of the user interface family of specs like XHTML. Why should it be in a specification all on its own?

 

If XLink isn't in the user interface space then I don't know what it's doing: I don't want standard attributes for defining relationships, I want to define my own.

 

I think XLink has never really decided whether it's in the "information content" space or the "user interface" space, and that's why no-one is using it.

    You are referring to the "xlink:show" attribute which clearly belongs in the UI space. Partially this is also true for the "xlink:actuate" attribute. You have to agree that this type of element has equal use in both "information content" space and "user interface" space. Are you suggesting that if the attributes "xlink:show" & "xlink:actuate" would be missing from the specification then the specification would be more successful ?
    If I understand correctly you don't agree with the UI space in the XLINK specification. What about the rest ? With XLINK you can define a full graph as links between resources (local & remote) and also document this process in a standard way. Not to mention the bidirectional links and the capability to define links to documents outside of your control.

    IMO XLINKS has failed because of 2 reasons:
    1. their power is not really needed in most applications;
    2. people are less informed about the capabilities of this specification; XML is already a sea of specifications; people are glad when they don't need to read about "YYY" because anybody is saying that "YYY" is not really useful anyway.