Lists Home |
Date Index |
This idea just keeps getting worse. You're basically asking for a centralized version of RDDL, the benefits of which are dubious at best but the negatiove ramifications could be significant.
From: Chiusano Joseph [mailto:email@example.com]
Sent: Wed 2/26/2003 4:27 PM
To: Jeff Lowery
Cc: XML Dev; firstname.lastname@example.org
Subject: Re: [xml-dev] Registered Namespace prefixes
[Cross-posting from XML-DEV to the OASIS/ebXML Registry TC]
Regarding the following:
My opinion is that the answer lies in a prefix registry. I know that's
controversial, mainly because it creates an authority structure that has
to be consulted prior to assigning prefixes to names. I think this can
be mitigate, though, by having a provisional namespace prefix mechanism
that is essentially the same as it exists now, minus default
namespaces. Registered prefixes would then be denoted by special naming
I think you are *very very* close to something, but not quite there.
What I mean is this: namespace prefixes in XML schemas/documents are the
proverbial "syntactic sugar" - that is, they are really "local" to the
XML schema/document in which they are declared. However, what really
matters is the actual namespace identifier that the prefix represents
(that is the "something" that you were close to). I believe it would be
very beneficial for one to be able to use a "namespace registry", so
that they could accurately reference namespace identifiers and "include"
them (using term loosely) in XML schemas/documents with whatever prefix
This opens up all sorts of possibilities in an XML registry, such as:
(1)Query on all "XML artifacts" (elements/attributes/datatypes" that
are in a given namespace;
(2)Reassign XML artifacts from one namespace to another (would
automatically search all registered XML schemas/documents for all
declarations/use of such artifacts in such namespace and change the
namespace identifier in the XML schema/document);
(3)Perform an automatic comparison of the "contents" of 2 (or more)
namespaces, for harmonization purposes (perhaps an organization has a
"test" namespace and an equivalent "production" namespace, and they wish
to determine which artifacts need to be promoted from test to production
at the proper time);
...and countless more.
Thanks so much for bringing this idea up.
Booz | Allen | Hamilton
Member, OASIS/ebXML Registry TC
Jeff Lowery wrote:
> Wait a sec while I change into my Nomex suit...
> I'd like to propose a mechanism for minimizing namespace hassles while
> maintaining readability. I expect this will raise hackles immediately, but
> hear me out:
> The mechanism for declaring namespace prefixes seems to be the primary
> failure point for namespaces. The association by scope of a prefix and it's
> declaration gives rise to all sorts of mischief when scope changes during
> document manipulation. Add default namespace declarations and things get way
> too interesting sometimes.
> All-in-all, given the design motivations of the WG, the basic mechanism is
> sound on a syntax level. Unfortunately, it creates dependencies withing a
> document that then need to be managed both internally and externally. Is
> there a way to manage these dependencies better, make them more
> My opinion is that the answer lies in a prefix registry. I know that's
> controversial, mainly because it creates an authority structure that has to
> be consulted prior to assigning prefixes to names. I think this can be
> mitigate, though, by having a provisional namespace prefix mechanism that is
> essentially the same as it exists now, minus default namespaces. Registered
> prefixes would then be denoted by special naming conventions.
> The advantage of a registry is that prefixed names become universal names
> when prefixes are registered. There are no scope issues. The primary
> disadvantage of registration is that there will be a prefix rush. I don't
> see a dependency on access to the registry at parse time, unless there are
> resources to be associated with the prefix (such as a URI to a RDDL doc)
> that the parser needs.
> And, lastly, default namespace declarations would have to go...
> I'm sure this is not a new proposal, but it's been at least a year since it
> was shot down last time... :-} Those permathreads need regular wear or
> they grow stiff.
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
> The list archives are at http://lists.xml.org/archives/xml-dev/
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://lists.xml.org/ob/adm.pl>