[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: XML Design for Diverse Data

• From: "Costello, Roger L." <costello@m...>
• To: <xml-dev@l...>
• Date: Fri, 25 Apr 2008 07:53:01 -0400

Hi Bruce,

> Even for compound documents, extension by the NVDL method 
> seems even scarier than the ANY element ... 

Actually, I see it just the opposite.  NVDL provides a "safe" way to
achieve extensibility.

The <any/> element represents a security risk (in certain
environments).

With NVDL we can have both extensibility and safety.  Here's how:

1. Each schema is designed in a "safe" fashion (i.e. without using the
<any/> element).  

2. Then, an XML instance document is created by assembling components,
each component conforming to a safe schema.  

3. The NVDL dispatcher ensures that an instance document contains just
the desired (safe) components, and nothing else.

4. An application processing the instance document processes some of
the components, or processes all the components, and in whatever order
makes sense to the application.

5. The application understands each component, because the schema for
each component has been examined and understood by the application
developer.  

6. The application understands the instance document, because the
meta-schema (i.e. the NVDL document) has been examined and understood
by the application developer.

Flexible, extensible, and safe ... Nice!

/Roger

• Follow-Ups:
  • Re: XML Design for Diverse Data
    • From: "bryan rasmussen" <rasmussen.bryan@g...>

• References:
  • Re: XML Design for Diverse Data
    • From: "bryan rasmussen" <rasmussen.bryan@g...>
  • RE: XML Design for Diverse Data
    • From: "Cox, Bruce" <Bruce.Cox@U...>