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

RE: RE: Declarative programming requires a different mindset

• From: "G. Ken Holman" <gkholman@CraneSoftwrights.com>
• To: "xml-dev@lists.xml.org" <xml-dev@lists.xml.org>
• Date: Sun, 11 Apr 2010 17:52:54 -0400

At 2010-04-11 16:29 -0400, Costello, Roger L. wrote:
>If XSD is not a programming language then what is it?
>
>People have said that XSD is:
>    - a declarative language
>    - descriptive markup
>
>I think that a programming language is a syntax for solving problems 
>(where the syntax is such that it can be processed by a computer).

Fine ... provided you use the plural "problems" ... I think a 
programming language is a syntax of semantic actions that can be 
arranged and rearranged by the writer to produce different actions on 
the same or different inputs, thus solving different kinds of problems.

>XSD is a syntax for solving certain problems. Specifically, XSD is 
>for solving constraint problems, i.e.,
>
>     Is the input ____ constrained in this ____ way?

I disagree.  That sounds too imperative to me.  Wouldn't "does the 
input violate any of these constraints?" be a better way of saying 
it?  I don't think that is simply the opposite.  How about "does the 
input satisfy these constraints?"?

All schema languages are declarations of constraints.  *How* a 
constraint checker checks the constraints is irrelevant to the fact 
that only one process is happening:  constraint-checking.  An XML 
document describing constraints isn't much different than an XML 
document describing a spreadsheet, or a book index, or anything other 
collection of information.

The semantics of the constraints are different with different languages:

   DTD - a pattern grammar
   Regular Language for XML (RELAX-NG) - a pattern grammar
   W3C Schema - a type hierarchy
   ISO Schematron - a set of assertions
   NVDL - one ring to bind them

>A computer can evaluate an input document against an XSD document 
>and produce an answer to the question, i.e., using the XSD language 
>the problem can be solved.

Again, I wouldn't word it that way.  The XSD language describes the 
constraints and a constraint checker checks to see if any constraints 
are violated.  It isn't changing the problem being solved.  It is 
just changing the inputs to the one unchanging problem being solved.

>I understand that the set of problems that can be solved using XSD 
>are very specific.

"set of problems"?  I disagree.  I see only one:  describe the 
constraints on elements, attributes and text in markup.  An XSD 
front-end is a process interpreting what those constraints are when 
massaging elements, text and attributes for application processing, 
but that doesn't make the XSD a programming language.

A W3C Schema expression doesn't describe a process.  It describes the 
end result of what a fixed and well-defined process described by the 
specification produces when acting on it.  If I want to change an 
application's view of <item>123456</item> from "123456" to the number 
123,456 then I change the inputs to change the type of constraint on 
the element.  But I haven't changed the problem.

The W3C Schema processor runs a process with the expression as 
input.  That process doesn't change based on the input, but what is 
checked does change based on the input.

>People have said that a programming language is one that can solve 
>many different kinds of problems. Certainly, then, XSD is not a 
>programming language.
>
>Both XSD and programming languages are languages for solving problems.

I suppose.  XSD solves *one* problem and a programming language can 
be used to solve many and varied problems.

>So, if I were to draw a Venn diagram I would draw a:
>
>    - bubble for XSD
>    - bubble for programming languages (independent of the XSD bubble)
>    - put both of the bubbles inside a big bubble
>
>What would you label the big bubble?
>
>Here's a graphic to illustrate my question:
>
>http://www.xfront.com/XML-Declarative-Programming/XSD-and-programming-languages-are-a-type-of-what.gif 
>

"The infinite set of XML instances"?  I think anything you come up 
with would be a tautology of "all things" based on the limited 
description of what you want.

Turn it around and I might have a better idea of what you are trying 
to look for ... what would you, Roger, use as an example of something 
*outside* of the big bubble?

Why have the big bubble?  What is missing in your work with XML that 
needs such a bubble to be defined?

I hope this helps.

. . . . . . . . . . . Ken


--
XSLT/XQuery training:         San Carlos, California 2010-04-26/30
Principles of XSLT for XQuery Writers: San Francisco,CA 2010-05-03
XSLT/XQuery training:                 Ottawa, Canada 2010-05-10/14
XSLT/XQuery/UBL/Code List training: Trondheim,Norway 2010-06-02/11
Vote for your XML training:   http://www.CraneSoftwrights.com/x/i/
Crane Softwrights Ltd.          http://www.CraneSoftwrights.com/x/
G. Ken Holman                 mailto:gkholman@CraneSoftwrights.com
Male Cancer Awareness Nov'07  http://www.CraneSoftwrights.com/x/bc
Legal business disclaimers:  http://www.CraneSoftwrights.com/legal

• References:
  • RE: Declarative programming requires a different mindset
    • From: "Costello, Roger L." <costello@mitre.org>
  • Re: RE: Declarative programming requires a differentmindset
    • From: Liam R E Quin <liam@w3.org>
  • RE: RE: Declarative programming requires a differentmindset
    • From: "Costello, Roger L." <costello@mitre.org>