RE: [xml-dev] Article: Keeping pace with James Clark

From
Bullard, Claude L (Len) <>
To
'Joe English' <>,
Date
2002-07-19T15:06:58Z
ID
<2C61CCE8A870D211A523080009B94E430752B6B5@HQ5>
Thread
RE: [xml-dev] Article: Keeping pace with James Clark
If you believe that, then you haven't been paying 
attention.  This issue is the black hole of WWW 
architecture.

See W3C TAG thread, 

"RE: httpRange-14 , what's the problem"

Fashion is, IMO, not the way to design interoperable 
systems.  

len


From: Joe English [mailto:]

Bullard, Claude L wrote:

> It is not healthy to be an ostrich among the coyotes.
>
> Is or is not a URI dereferenceable?  If you say,
> it depends, you have your head in the sand.

I would say that a URL is dereferencable, unequivocably.
It locates a resource.

But URLs are out of fashion.  The current mode
is to Identify resources, not Locate them.  URIs are
used in several W3C specs -- RDF and XMLNS being the
most prominent -- as opaque, atomic identifiers
that have no semantics other than identity.

If you say that a URI must be dereferencable just
because it looks like a URL, then you haven't been
paying attention.