RE: The Web as an Application

David,

> 
> Good point I forgot that base was for xlink not external entities.
> Still this model suggests that hypermedia should be at the 
> same level as xlink in terms of specifications.

Well that depends on how you interpret it.  By my interpretation, 
links are so important that they need to be available to other XML
application, for the identical reason that xml:base is so designed,
and in the identical manner.

I think TimBL agrees with the importance of links, somehow.  I am reading his 
book "Weaving The Web" now, and although I don't have it at hand,
he says effectively that everything can be thought of in terms of
its linkages to other things.

XML is designed around relationships between things, by virtue of the
nesting of elements and attributes.  This is invariant across XML.

I argue that links between documents are as fundamental as nesting
of elements and attributes, and so merit a place in the xml namespace 
i.e. invariant across all xml.  That would have included MicroXML too, 
had they not ditched the xml: namespace, too.  Too bad for them, I think.

> 
> ( argue why xlink is not in core XML ?)

Bad design choice.  But a lucky one, since xlink is so complicated it
doesn't get used.

> 
> Furthermore I suggest for practical purposes the different is 
> moot.  I have extrmely confidence ANYONE can get more stuff 
> added to core XML unless it is shown to have high adoption, 
> Which is a chicken and egg problem ... But it is still reality.

Yes, rough consensus and running code should rule the day.

I would love to see an XML parser modified to support xml:href et al.,
I wonder if it could be easily done with SAX, say.  And what benefit
it might confer on applications that used it.   


> 
> I suggest for practical reasons if we want to make progress 
> we examine xlink and determine why exactly it fails to meet 
> the needs of hypermedia, and if so propose a Change or 
> replacement to that.

I have Erik Wilde and David Lowe's book on XLink and shall pull some info out of there.

Possibly we should devote a wiki page or two to this topic too.

Cheers,
Peter

ps having enormous difficulties with Outlook today, so may not respond again for
a while...


> 
> 
> 
> Sent from my iPad (excuse the terseness) David A Lee 
> dlee@marklogic.com
> 812-630-7622
> 
> 
> On Jun 6, 2013, at 12:59 PM, "Rushforth, Peter" 
> <Peter.Rushforth@NRCan-RNCan.gc.ca> wrote:
> 
> > David,
> > 
> >> Only one of these (xml:base) is even close to language 
> >> oriented.   and xml:base is *only* there to serve the problem 
> >> which was generated by allowing external parsed entities.   
> >> xml:base doesnt let you link to anything ... it is only 
> >> informational about where parts of a XML docuement may have 
> >> come from that were not where you originally asked for it.   
> > 
> > To quote from [1]: 
> > 
> > "This document describes a mechanism for 
> > providing base URI services to XLink, but as a modular 
> > specification so that other XML applications benefiting 
> > from additional control over relative URIs but not built upon 
> > XLink can also make use of it."
> > [1] http://www.w3.org/TR/xmlbase/
> > 
> >> It has nothing at all to do with the application layer.
> > 
> > Not true.  See above reference to "other XML applications". 
> > 
> > And the spirit of pipe and filter seems to me reflected in that
> > statement.  And many applications could benefit from 'control
> > over relative URIs', although the important fact is URI, 
> > relative or absolute.  
> > 
> > Cheers,
> > Peter
>

Received on Thursday, 6 June 2013 18:14:11 UTC