words (RE: extensibility in XSchema?)

[Paul Prescod]

On Mon, 22 Jun 1998, Peter Murray-Rust wrote:
> 
> I'll buy this - at least to the extent that I shall try to avoid the term
> semantics.  However I need *some* terminology to describe what I want to
> do. Perhaps we can converge on some of that (given also that spec drafters
> are also trying to firm up *some* of the terminological problems). I shall
> use the word 'meaning' at present.

I think that the word "semantics" is fine.

> - how do I add human 'meaning' to an object (e.g. an element or attribute)?

Documentation, lectures, email, lunch discussions.

> - how do I add machine 'meaning' to an object?

Computers aren't yet smart enough to understand anything. It will be 
decades before you will be able to describe the concept of "title" to a 
computer. All we can teach computers is *behaviour*: what to do with 
various types of XML elements.

> At present there are *no* defined mechanisms for either of these in XML. I
> find this amazing and regrettable (I - and others such have Rick) have been
> shouting for it).  

Well, the first is easy. I would be in favour of having the namespaces 
spec. require the namespace PI point to documentation for a namespace. 
Otherwise, we could easily develop another PI.

> Now compare MIME:
> 	If I get sent a file of type image/gif I can:
> 	- determine that 'image/gif' is a MIME type
> 	- see if it is in the central IETF registry. If so I can read the formal
> definition, etc.

Not all MIME types have formal definitions.

> 	- *and most important for me* I can associate the MIME type with a 
> program through the mailcap file, *even if the MIME type is not 
> registered*.
> This is all I am asking for, but there seems to be no way to do a similar
> thing for an XML element or attribute. 

What would the program do? Print it? Search it? Convert it to another 
format? If we restrict our discussion to printing: would it print it in 
batch mode, or present a WYSIWYG view of it? would it print landscape or 
portrait?

There are an infinite number of interesting behaviours that go with each 
element type. How would the mailcap handle that?

> This is what I used to call semantics. If someone can come up with a better
> term, it would be useful. RickJ uses the word 'bind' - I like that.

Semantics are fine for human meaning. "Behaviour" seems okay for the 
rest. Note that I even got the spelling correct.

 Paul Prescod


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)