[Home] [By Thread] [By Date] [Recent Entries]

  • From: Michael Sokolov <msokolov@s...>
  • To: Uche Ogbuji <uche@o...>, Pete Cordell <petexmldev@c...>
  • Date: Fri, 15 Nov 2013 07:49:17 -0500

On 11/14/2013 10:36 AM, Uche Ogbuji wrote:
On Thu, Nov 14, 2013 at 8:31 AM, Pete Cordell <petexmldev@c...> wrote:
It's ironic that the design of associating a namespace with every element meant that you didn't have to drag context information from parent nodes (as would be required in Michael's example below), but the implementation using prefixes meant that you have to drag context information from parent nodes!

Well yes. Even though making URIs an intrinsic part of names was a mistake in XML Namespaces, it wasn't the biggest mistake. The biggest mistake was the hierarchical scoping mechanism that spawned prefixes, and thus the ultimate evil of QNames in content.
I'm also not fond of a data model that defines a name as a three-part entity (prefix, localname, namespace) instead of a single string: it's a mess.

-Mike


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]


Site Map | Privacy Policy | Terms of Use | Trademarks
Free Stylus Studio XML Training:
W3C Member