Cart
[Home] [By Thread] [By Date] [Recent Entries]
Paul Spencer wrote: > > But now RDDL comes along and gives us something for the namespace to point > to. This could result in a need to version namespaces so that we can point > to different RDDL documents. But if we do this, what happens (as David > Carlisle mentioned) to processing code that uses fully-qualified names (such > as code that uses XPath expressions)? I'm just starting to think about this > for UK Govt. > I think the reasons whether or not to change a namespace URI (i.e. create a new namespace) when new versions of something come out have been well addressed. Generally I agree that changes in a namespace vocabulary which _are not_ backward compatible should be accompanied by a new namespace, whereas version changes that _are_ backward compatible should use the same namespace (in order to allow software to continue to recognize the documents). These are all important design decisions. RDDL does provide a mechanism to tie versioning of namespace related schemata to the namespace. One can embed multiple rddl:resources with natures and purposes that indicate prior versions of the same type of resource (e.g. XML Schema). Similarly a RDDL resource might indicate a related namespace URI, perhaps nature="http://www.rddl.org/" and purpose="http://www.rddl.org/purposes#prior-version" would indicate a prior version of a RDDL document/namespace URI , whereas the purpose http://www.rddl.org/purposes#prior-version with a nature http://www.w3.org/2001/XMLSchema would indicate a prior version of an XML Schema in the same namespace. Does this make sense? -Jonathan
|
