W3C home > Mailing lists > Public > public-html@w3.org > August 2007

Re: Distributed Extensibility

From: Henri Sivonen <hsivonen@iki.fi>
Date: Mon, 6 Aug 2007 18:17:07 +0300
Message-Id: <497ACCC7-6B4D-409E-854F-C076D5A89BCA@iki.fi>
Cc: public-html@w3.org
To: Sam Ruby <rubys@us.ibm.com>

On Aug 6, 2007, at 15:14, Sam Ruby wrote:

> Consider the following XHTML fragment:
> <a:b xmlns:a='urn'/>
> Questions:

These questions allude to the goal being enabling XML namespaces in  
text/html rather than coming up with *some* method of reducing the  
probability of extension name collisions. There's a tradeoff between  
backwards compatibility and using XML namespaces as the collision  
avoidance mechanism.

Philip's suggestion involved the colon but didn't go to namespaces  
all the way. My point is that if we decide to do extensibility at all  
with some kind of extender-assigned qualifiers, we should either
  1) Not use the colon, keep the extensions in the XHTML namespace in  
both serializations and be backwards compatible.
  2) Go all the way to an alternative serialization for namespaced  
XML 1.0 at the expense of backward compatibility.

Going half-way with the colon would make the XML and text/html  
serializations diverge for good.

Doing in-browser namespaced DOM nodes with the motivation bringing  
existing XML languages to text/html (as opposed to enabling yet-to-be  
created extensions) carries different considerations.

Henri Sivonen
Received on Monday, 6 August 2007 15:17:25 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:25 UTC