W3C home > Mailing lists > Public > www-webdav-dasl@w3.org > April to June 1998

RE: let's subset XML Data

From: Jim Davis <jdavis@parc.xerox.com>
Date: Sun, 21 Jun 1998 14:46:28 PDT
Message-Id: <3.0.3.32.19980621144628.007fa990@mailback.parc.xerox.com>
To: www-webdav-dasl@w3.org
At 12:51 AM 6/21/98 PDT, Saveen Reddy (Exchange) wrote:
>Why is having a hardwired URN a compatibility risk? This is no more of a
>compatibility risk than using the "DAV:" namespace (or namespaces in
>general).

Au contraire.  The Namespace PI allows you to specify the URN in the XML
processing instruction, so if you decide one day to use a different URN
(e.g. DAV2:) you just write the XML and you are done. On the other hand,
XML Data mandates a *hidden* URN that is to be used by default.

If you write, e.g. <foo dt:dt="int"> as opposed to <foo dt:dt="dt:int">
then the application must know the URN to substitute into the attribute
value.   If, someday, you want to switch to some new definition of int, or
if, in the process of standardizing XML Data the default URN changes you
are hosed.  Well, you say, no you are not hosed, you just start using an
explicit URN at that time.  Yeah right I say, so let's start writing
explicit URNs now, and not wait.  It's only two extra characters, and it's
safer.
Received on Sunday, 21 June 1998 17:48:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 22 March 2009 03:38:04 GMT