Re: (Second) Last Call for XPointer 1.0

[Jonathan Borden <jborden@mediaone.net>]

Tim Bray wrote:

> At 09:35 AM 09/01/01 -0500, Elliotte Rusty Harold wrote:
> > It's
> >also come to light in this draft that Sun claims a patent on some of
> >the technologies needed to implement XPointer.
> ...
> >I think this is particularly offensive
>
> Offensive or not, it's a big problem.
>
...
>
> XPointer, D.O.A.


Yep.

Perhaps there's a fix:

"XPointer is built on top of the XML Path Language [XPath], which is an
expression language underlying the XSL Transformations (XSLT) language.
XPointer's extensions to XPath allow it to:

Address points and ranges as well as whole nodes
Locate information by string matching
Use addressing expressions in URI references as fragment identifiers (after
suitable escaping)
"

It appears the Sun patent applies (perhaps not at all, but if so) only to
the second item: "Locate information by string matching".

Loose string matching if that's what it takes to save XPointer.

The other issue is XPointer as the text/xml fragment identifier. As part of
a URI reference, the format of fragment identifiers is defined on a
media-type basis, in this case "text/xml" or "application/xml". (per IETF
RFC 2396). XPointer should not be certified as the XML fragment identifier
format until it can be used without restrictions. Doesn't the IETF tend to
frown on encumbered 'standards'?

-Jonathan