Re: Namespace URLs (was RE: [Big-devel] big bad mother schema)

[Jim C. <jr...@an...>]

>The URLs I'm referring to are not for general use.  They are intended as
>the source for a particular version of a particular data standard set in
>XML Schema.

well they might as well be.  Then they can serve the function of the 
namespace definition.  The is no point in being secretive about these things...

>In the case of XML Schema, the URL we use needs to point to a *.xsd
>document.  As this document eventually becomes the reference for a lot
>of applications (such as XML Spy), they come to expect to find a
>particular version of a schema at that location.
>The URL must be unique to that schema version.

yes...

>If we change this URL, we potentially break any applications making use
>of the standard (because of incompatibility between versions, for
>example).  This tends to annoy application developers somewhat.

yes - once it is in place, that is where and how it has to stay... this 
might be a problem with souurceforge, but not with chah...

>Thus, we must be careful to provide a unique but guessable schema
>location URL for each version of HISPID that we choose to develop in XML
>Schema (or whatever comes next..).

machines do not guess... all it has to be is unambiguous...  obviously 
simpler is better though...

jim


~ Jim Croft ~ jr...@an... ~ 02-62465500 ~ www.anbg.gov.au/jrc/ ~