RE: [xml-dev] (In)Validate My Assumptions on Linking.

["Nathan Young -X \(natyoung - Artizen at Cisco\)" <natyoung@cisco.com>]

I had a busy week so didn't get to reply to this in a timely way.

Many people seem to have ideas about linking use cases... xlink was
implemented to handle some of them, XHTML some more, topic maps some as
well... RDF... I know not every implementation is working to support the
same set.  Would it help to enumerate the use cases we are talking about
when we say "links"??

For example:

 1 unidirectional link from a section in a document to another document
 2 unidirectional link from a section of a document to another point in
the same document
 3 unidirectional link from a section of a document to a point in
another document
 4 document composed of several other documents (but with no other
content in the document except references to the composing docs)
 5 document with real content with some additional content coming from
other documents

These cases are supported to some degree by XHTML.

1-3 are supported by anchor tags, with the caveats:
 - the section that the link comes from must span a well formed section
of the document (and there are other restrictions on what an anchor can
contain)
 - for cases 2 and 3 the destination points must be defined in the
destination document.  You could argue that XHTML supports pointing to
sections rather than points but the browsers don't.

4 and 5 are supported in XHTML by framesets and iframes respectively. In
both cases there are significant structural and layout restrictions that
web authors will be well aware of.

Here are some more:

 6 link from a point in a document to code that runs when the link is
encountered
 7 link from a document to another document with style information in it
 8 link from a point in a document to several other documents (or points
within those documents)
 9 links as meta data about a document that indicate that it is relevant
to another document or section thereof

6 and 7 are also supported by XHTML.  8 can be implemented in javascript
as Ben pointed out, but the methods for representing it and implementing
it are not standardized. 9 is implemented as annotations (see the
annotea project).

I feel like I'm barely scraping the surface here and I've mainly focused
on cases that XHTML supports to some degree.

I'd be interested in contributing to a more exhaustive list, maybe in a
wiki somewhere?  Or enumerated in this thread?  Ben are you maintaining
anything like this?

------------>Nathan
 





.:||:._.:||:._.:||:._.:||:._.:||:._.:||:._.:||:._.:||:._.:||:._.:||:._.:
||:.

Nathan Young
Cisco.com->Interface Development
A: ncy1717
E: natyoung@cisco.com  

> -----Original Message-----
> From: peter murray-rust [mailto:pm286@cam.ac.uk] 
> Sent: Thursday, September 28, 2006 11:50 PM
> To: xml-dev@lists.xml.org
> Subject: Re: [xml-dev] (In)Validate My Assumptions on Linking.
> 
> At 19:55 28/09/2006, Ben Trafford wrote:
> 
> >At 02:44 PM 9/28/2006, Len Bullard wrote:
> >>As I said elsewhere, look at ISO 10744 which is about as 
> exhaustive a work
> >>as I've encountered, but it isn't easy sledding, so look at 
> what Elliot
> >>wrote about it.
> >
> >         Read it. All of it. For the third time. I like HyTime, a 
> > lot. It informed XLink development on a number of levels, and by 
> > proxy, will inform my current efforts.
> 
> I am sure Ben has it covered but two of the important concepts (or 
> perhaps one) are the compound document and the hyperdocument (which I 
> think is the same as a bounded object set). In other words I want to 
> be able to create a set of links that define an information object 
> and send those links to a human or machine. Typical examples are an 
> XHTML page with embedded objects (images, maths, chemistry) - is this 
> a compound document? and a complete directory or other 
> structure (BOS).
> 
> Obviously this requires something way beyond CSS...
> 
> P.
> 
> Again I compliment ben on his energy and commitment. This will not be 
> easy but it will be immensely valuable. I suspect it will be useful 
> to have some regular offerings even if these don't cover all the 
> aspects. And we certainly won't all get everything we want.
> 
> 
> (The Subjects appear to have got forked so I am resetting this one)
> 
> 
> 
> 
> Peter Murray-Rust
> Unilever Centre for Molecular Sciences Informatics
> University of Cambridge,
> Lensfield Road,  Cambridge CB2 1EW, UK
> +44-1223-763069 
> 
> 
> ______________________________________________________________
> _________
> 
> XML-DEV is a publicly archived, unmoderated list hosted by OASIS
> to support XML implementation and development. To minimize
> spam in the archives, you must subscribe before posting.
> 
> [Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
> Or unsubscribe: xml-dev-unsubscribe@lists.xml.org
> subscribe: xml-dev-subscribe@lists.xml.org
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
>