Re: URI for XML schema and namespace

Peter Saint-Andre <> Fri, 04 January 2008 16:20 UTC

Return-path: <>
Received: from [] ( by with esmtp (Exim 4.43) id 1JApHO-000346-0I; Fri, 04 Jan 2008 11:20:18 -0500
Received: from discuss by with local (Exim 4.43) id 1JApHM-00033x-5q for; Fri, 04 Jan 2008 11:20:16 -0500
Received: from [] ( by with esmtp (Exim 4.43) id 1JApHL-00033p-SB for; Fri, 04 Jan 2008 11:20:15 -0500
Received: from ([]) by with esmtp (Exim 4.43) id 1JApHL-0007B0-6Z for; Fri, 04 Jan 2008 11:20:15 -0500
Received: from ( []) (Authenticated sender: stpeter) by (Postfix) with ESMTP id A19C3404A1; Fri, 4 Jan 2008 09:20:08 -0700 (MST)
Message-ID: <>
Date: Fri, 04 Jan 2008 09:20:06 -0700
From: Peter Saint-Andre <>
User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv: Gecko/20071031 Thunderbird/ Mnenhy/
MIME-Version: 1.0
Subject: Re: URI for XML schema and namespace
References: <> <00a601c84edd$2ca4dcc0$0601a8c0@pc6> <000001c84ee8$789307a0$0223520a@charger> <>
In-Reply-To: <>
Content-Type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary="------------ms080309090907070208060707"
X-Spam-Score: -0.0 (/)
X-Scan-Signature: 97c820c82c68af374c4e382a80dc5017
Cc: Apps Discuss <>
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: general discussion of application-layer protocols <>
List-Unsubscribe: <>, <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>

Tim Bray wrote:
> On Jan 4, 2008 7:42 AM, Scott Hollenbeck <> wrote:
>>> What is the currently recommended form for a URI for an XML
>>> schema and an XML namespace?
>>> The question has come up on ForCES where I, citing RFC3688
>>> say it should be urn:
>>> whereas the others cite RFC3470 to say that http: is recommended.
>>> Other opinions welcome.
> Well, the Architecture of the World Wide Web
> ( argues strongly that http: URIs are
> generally useful on the grounds that they can in principle (don't have
> to be, but can) be used to retrieve some human-readable explanatory
> material about whatever it is the URI names.   Also the URI itself
> tends to be a bit more human-readable, which seems worthwhile to me.

If http URI, type into browser location bar.

If URN, type into browser search bar.

I don't see a great deal of difference. :)

> The downside, as Scott points out, is that if you want to do that, you
> have to find a domain to put after the http:// part, 

... and ensure that the domain is stable, and that the people running 
the domain will reliably host your namespace, and ...

> and the IETF just
> doesn't do that.  In the case of RFC4287 and 5023, we borrowed one
> from the W3C.

FWIW, we Jabberites used https URIs like* 
for a long time but recently switched to URNs like urn:xmpp:* because 
they're more stable (what happens if the person registering 
loses control over the domain?) and will last forever.

The choice of http URI vs. URN seems like a toss-up to me.


Peter Saint-Andre