----- Original Message -----
From: "Stephane Bortzmeyer" <bortzmeyer(_at_)nic(_dot_)fr>
To: "Tom.Petch" <sisyphus(_at_)dial(_dot_)pipex(_dot_)com>
Cc: "ietf" <ietf(_at_)ietf(_dot_)org>; <simple(_at_)ietf(_dot_)org>
Sent: Tuesday, September 18, 2007 4:10 PM
Subject: Re: Last Call: draft-ietf-simple-xml-patch-ops (An Extensible Markup
Language (XML) Patch Operations Framework Utilizing XML Path Language (XPath)
Selectors) to Proposed Standard
On Tue, Sep 18, 2007 at 02:19:51PM +0200,
Tom.Petch <sisyphus(_at_)dial(_dot_)pipex(_dot_)com> wrote
a message of 135 lines which said:
I question the use of XPath 1.0, when XPath 2.0 was approved at the
start of this year. It seems short-sighted, a bit like choosing IPv4
over IPv6
I strongly disagree. Xpath 2.0 is *much* more complicated than Xpath
1.0. Among free software, there is little implementation (or even
plans) of 2.0. Xpath 2.0 is quite controversial.
OK, XPath 2.0 looks more attractive to me as a user but I am not familiar with
the uptake of XPath 2.0 (and does this make
my point that more IETF-wide guidance could be valuable eg, to set another pack
of leverets running, XSD v RELAX NG v ...)
The comparison with IPv4/v6 is wrong. If you start from scratch, IPv6
is no more complicated than IPv4 (and it is probably the
opposite). Xpath 2.0 is always much more difficult to implement (for
instance, it requires schemas).
This business of updating parts of an XML document seems to be
cropping up in a number of places in the IETF with very different
solutions.
AFAIK, this is the first one to be specified at IETF. Other contenders
are:
* REX (W3C), which uses DOM events http://www.w3.org/TR/rex/
* Xquery update (W3C) http://www.w3.org/TR/xqupdate/
* XUpdate, which seems completely dead
http://xmldb-org.sourceforge.net/xupdate/
* DUL, there was an I-D, "A delta format for XML documents",
draft-mouat-xml-patch-00.txt, now expired
http://sourceforge.net/projects/diffxml
Two that are current for me NETCONF, which offers XPath or subtree filtering to
update configuration information stored in XML, and ForCES which uses a
different approach to configure Network Elements specified via XML; I think I
have seen others in the Apps area. I am not saying that these are doing exactly
the same, rather that I see the same issues with varying results.
Tom Petch
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf