RE: [xml-dev] XML and the Relational Model (was Re: [xml-dev] A standard approach to glueing together ...)

["Hunsberger, Peter" <Peter.Hunsberger@stjude.org>]

Thomas B. Passin <tpassin@comcast.net> asks:
> 
> [Hunsberger, Peter]
> 
> Thomas B. Passin <tpassin@comcast.net> asks (among many other things):
> 
> > * Are there things in the XML world that are equivalent to 
> relational 
> > database views, and if so what are they and do they have any 
> > advantages? Will xquery change this picture? Would view equivalents 
> > make updates easier or harder?
> 
> I'd propose that an XPath statement in general can be 
> considered the XML equivalent to a view...
> 
> Well, that works at least to some extent, given that a view 
> is generally based on an SQL select statement, and a 
> "select='{xpath_expression_here}' is something like.  What 
> about xslt expressions that go beyond what you can cram into 
> a single  xpath expression?  What about a whole stylesheet?  
> If some stylesheet results could be considered equivalent to 
> a view, which ones cannot?

Well, in the SQL world there are clearly some data extractions that
cannot be obtained from a view.  To me, a transform seems more akin to a
stored procedure than a view?

> Let's see, a view can combine data atoms from different 
> tables, based on some join conditions.  Could an XML view 
> could combine infoset items from different documents, or just 
> from different elements within a single document?

An SQL view can join across multiple database instances in some
implementations.  But then again, XPath with the document function
works, so I see no need to generalize to XQuery (again you seem to be
picking up some stored procedure type capabilities).  However, this is
one set of analogies I'm not going to be very picky about.  I like
Xpath, and don't have experience with Xquery, so for the moment Xpath is
my golden hammer...