W3C home > Mailing lists > Public > xml-dist-app@w3.org > December 2001

Re: NameValue and NameValueList data types

From: Pete Hendry <peter.hendry@capeclear.com>
Date: Fri, 21 Dec 2001 12:54:04 +0000
Message-ID: <3C2330EC.4010908@capeclear.com>
To: Rich Salz <rsalz@zolera.com>
CC: xml-dist-app@w3.org

 > You have not responded to the counter-arguments that "array of structs
 > containing (key, value) is good enough," and why you feel it is not okay
 > to have that on a per-schema basis.

And int, short, etc. could be defined in every schema deriving from integer so why have them?

> In fact, the very point that you can
> define a KeyValue list on a per-schema basis can be seen as a proof that
> it's not a basic type.


Your assertion that this is proof that it is not a basic type would imply that any derived type is not basic as it can 
be derived in every schema. The basic types would then be all the types in schema which are derived directly from 
anyType or anySimpleType and no others!

For me a basic type from which other types are built and/or it is defined in (almost) every programming language.

It is not KeyValueList that I am proposing (although it was originally, but after some thought) it is only KeyValue. 
Just as int is a basic type but int[] is not, so KeyValue is but KeyValue[] is not.

Pete
Received on Friday, 21 December 2001 07:56:01 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 22:28:13 UTC