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

RE: The limitations of XPath and navigation for XML

• From: "Michael Kay" <mike@s...>
• To: <mike@a...>,"'Xia Li'" <xli@g...>,<noah_mendelsohn@u...>
• Date: Fri, 8 Feb 2008 10:09:17 -0000

			
> What about ?//C//G?.  To me as an external user, it says go to C, then go
to G (from C).

We tend to create mental models of how computers work, and those models are
often procedural. For example, my mental model of 

  SELECT * FROM EMP WHERE EMP.SALARY > 10000

is that it scans the EMP table testing each row against the predicate and
outputting it if the predicate is true. But the fact that I have this mental
model does not make SQL procedural. That's true, incidentally, regardless of
whether my mental model bears any resemblance to real implementations (in
both these cases, the model is sometimes accurate and sometimes not).

Michael Kay
http://www.saxonica.com/
			

• References:
  • Re: The limitations of XPath and navigation for XML database processing
    • From: mike@a...