RE: Hypermedia - Why

David,

> -----Original Message-----
> From: David Carlisle [mailto:davidc@nag.co.uk] 
> Sent: July 25, 2012 15:27
> To: public-xmlhypermedia@w3.org
> Subject: Re: Hypermedia - Why
> 
> I wrote
> 
> > I am asking why? Given all the evidence I see is that language 
> > designers would not use them even if they were defined.

Well, first the hypermedia community is engaged with this effort,  so I think
it best to ask them if they would use such devices, if defined.

> 
> On 25/07/2012 17:30, Rushforth, Peter wrote:
> > These attributes haven't been proposed before
> 
> Maybe I missed something but apart from some details about 
> specific attributes the proposal is to put xlink style 
> attributes in the xml rather than xlink namespace, so most of 
> the issues are shared with xlink.
> 
> and xlink at one point used the xml namespace eg
> 
> http://www.w3.org/TR/1998/WD-xlink-19980303
> 
> defined xml:link but during development the decision was 
> taken to use an xlink specific namespace.


As to "Why are hypermedia affordance vowels necessary at this time", I think the answer has a few
facets.

First, the ostensible goal of XML is to be successful on the internet.  Now there are a few caveats
to that, see some of the annotations linked to from the wiki page.  But overall, I believe that
goal can be taken to mean (something close to): "XML should be easy to transmit and receive using Web infrastructure".

The XLink effort, undertaken by the best and brightest of the SGML world, seems to have overshot the
runway.  I do not want to knock the efforts of that specification, but it does have a couple of 
issues.  The first is that it may be too complex  - designed for I'm not quite sure what system, but not the Web.
The one vowel in that vocabulary that seems to be worthwhile on the web is xlink:href.  I have not seen the
others used in practice unless forced to by a schema.  I hope Liam or Norm can point out some
systems where they are in use.  Maybe those systems should be or are using SGML!

I have seen you tell others on public lists that xlink is wrong, but I didn't challenge you as to why
you think it's wrong but I would like to know, without starting a flame war.  But I think there are
issues with it, else why would more modern and evolving vocabularies such as xinclude not rely on it.
Not to mention MathML.

So, it may be worthwhile to redesign a "new" "xlink" that uses the hypermedia vowels proposed.  That's one element of "why".

As to "why the xml namespace" aspect, I think it is important to hand developers the tools for the
job.  Namespaces are not part of XML, and indeed not everyone wants to use them, and I think the 
goal of XML is to be successful on the internet, namespaces or not.
To do that, I believe a language needs to have elements which reflect the environment of deployment.
It seems clear to me that web-oriented hypermedia affordances are getting re-invented in every vocabulary
of xml that gets minted.  So one way to encourage developers not to do that is to give them the tools
to avoid that problem.  The xml: namespace seems the only way.  If they would not be used, that
would be a "problem".  So I think you have to ask the hypermedia community, who we have on the line now.

Peter

Received on Thursday, 26 July 2012 12:28:50 UTC