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

Datatypes - it's in the contract

• To: xml-dev@l...
• Subject: Datatypes - it's in the contract
• From: Jonathan Robie <jonathan.robie@d...>
• Date: Sat, 28 Sep 2002 07:50:50 -0400
• In-reply-to: <5.1.0.14.0.20020928072413.05180ec8@n...>
• References: <200209280146.g8S1kR319235@d...><5.1.0.14.0.20020927194520.00bcae40@n...><5.1.0.14.0.20020927194520.00bcae40@n...><5.1.0.14.0.20020927180018.04398a18@n...>

To me, the most fundamental reason for datatypes is to allow them to be 
part of the contract between producers and consumers of data. Suppose we 
have a number of systems that work with data about people. Simon has 
suggested that we should specify only the names and nesting of elements and 
attributes. Consider the following example:

   <person>
      <ssn>123456789</ssn>
      <name>Jonathan Robie</name>
      <children>3</children>
   </person>

Using an XML 1.0 validator, if the above example is valid, so is the one below:

   <person>
      <ssn></ssn>
      <name>Jonathan Robie</name>
      <children>Hickory dickory dock</children>
   </person>

If the applications that use this data require data of the appropriate 
type, and we want validation to be able to determine whether the contract 
is being followed, then we have to allow data types to be declared.

Jonathan

• Follow-Ups:
  • Re: Datatypes - it's in the contract
    • From: Mike Champion <mc@x...>
  • Re: Datatypes - it's in the contract
    • From: Arjun Ray <aray@n...>
  • Re: Datatypes - it's in the contract
    • From: Eric van der Vlist <vdv@d...>

• References:
  • Re: limits of the generic
    • From: Joe English <jenglish@f...>
  • Re: limits of the generic
    • From: Jonathan Robie <jonathan.robie@d...>
  • Re: limits of the generic
    • From: Jonathan Robie <jonathan.robie@d...>
  • Re: limits of the generic
    • From: Jonathan Robie <jonathan.robie@d...>

• Prev by Date: Re: limits of the generic
• Next by Date: namespaces in XML 1.1 - namespace names become *IRI* references
• Previous by thread: Re: limits of the generic
• Next by thread: Re: Datatypes - it's in the contract
• Index(es):
  • Date
  • Thread