Re: Are *relative* URIs as namespace nemes considered harmful?

On Thu, 18 May 2000, Tim Berners-Lee wrote:
> XPath explicitly calls for absolutization before comparison.  

Gosh.  I have read the spec many times... and have
consistently missed this sentance in the specification.
I feel very dumb at the moment.    I guess I had never
expected the XPath spec to *redefine* the specific 
definitions (of identity) laid out so clearly in the
namespace specification.

Luckly, I don't think I am alone.  I tried out relative
namespaces on XT -- James Clark's software does not
seem to be doing this absolutization; but perhaps my
testing is such that I don't quite understand what
would reasonably be absolutized or how it would be
absolutized. Does *any* XSLT processor (implementing 
XPath) support relative URI absolutization?

Slightly Shocked and a tad bit Confused,

Clark

Received on Thursday, 18 May 2000 23:01:15 UTC