RE: [xml-dev] "Introducing MicroXML, Part 1: Explore the basicprinciples

["Rushforth, Peter" <Peter.Rushforth@NRCan-RNCan.gc.ca>]

Hi Len,

Thanks for chiming in.  It's going to be hard for me to reply to 
everything, I think, and probably my answers won't really be up to
the standard required for this discussion.  So everyone, please, add your
opinion, if you think it worthwhile and you understand the point.

> The disagreement originally was is hyperlinking owned by the 
> content or by the style sheet (say the application).  This is 
> the original Hytime vs DSSSL schism.  At that time, there was 
> no "web" to talk about.
> 
> IOW:  what IS a hyperlink?  A relationship?  A function? An object?
> What?

Who cares *what* it is.  It is used as a mechanism to transfer state
on the web.  In any case, what it is can be described by hints (@xml:rel etc) to the
application.  All the underlying infrastructure has to do is the fetching
and putting.  Probably security is an issue as to why the links 
should be recognizable.

> 
> Be careful of the "giants" and "REST is THE" discussion.   There is a
> bit of retrofitted history in that description.  The Internet 
> and the web applications as well as others were functional 
> before the term "representational state transfer" was ever 
> coined and that notion was already a part of discussions.

I realized that.  That's the great thing about the web, those giants can
communicate with us easily if they wish, as John's rotating tag line says
so neatly.  I just am saying that
the objective of XML was the Web, not 3D gaming, unless 3D gaming is
done on the web.  If the simple semantics which I've proposed don't fit
the use case for the 3D gaming application, invent your own which do.  But
the rule of least power, known in my level of the atmosphere as KISS, 
says that we should follow the 80-20 rule here.
  
> There is less invention there than you believe which is why 
> the W3C has been unable to overturn certain critical patents. 

Are you saying that XML is patented, and as a result we don't control that
namespace?  Maybe another namespace could be declared and used.

> However the technical point is as David and others have said. 
>  As soon as you add application semantics, XML becomes less 
> useful, not more.
> For example, try putting hard wired linking semantics into 
> non-text browser applications such as 3D gaming.

Hence the effort to toss stuff overboard.  But, the application, the Web,
is a pretty big use case.  I think HTML is big because of the reliable
hypermedia affordances.  But everyone is trying to extend the semantics
of HTML with what have you, yet the media type remains text/html. That is
the proper domain of xml.

Peter