D
Dave Dubin
XML Processing in Prolog: A tutorial at Extreme Markup Languages 2005
August 1, 8:30 AM-5:30 PM, Hotel Europa, Montreal, Canada
The XML of Programming Languages
The great appeal of XML for representing documents, business and
scientific data has its basis in a particular philosophy for
organizing work. As an XML user, you already understand the power of
describing things in terms of what they are, rather than how they look
or behave. You appreciate the flexibility gained by decoupling
definitions and descriptions from the details of how those data are
expected to be processed.
What would 'the XML of programming languages' be like? One answer,
of course, is that such a language would be an XML application itself,
capable of being processed just like (or even along with) other XML
markup.
This tutorial invites you to consider another answer: instead of a
programming language expressed *as* XML, we'll introduce one designed
around similar principles to those of declarative markup. Prolog was
developed at the Universities of Aix-Marseille and Edinburgh in the
early 1970s -- around the same time Charles Goldfarb was inventing
SGML at IBM. Colmerauer, Kowalski, and Roussel originally designed
Prolog to support research in natural language processing. It should
therefore come as no surprise that Prolog offers many benefits for
processing artificial languages, such as those expressed in XML.
Processing XML in Prolog
This tutorial will be an introduction to logic programming and to
problems, issues and strategies for XML processing in the Prolog
language. It is intended for experienced programmers who are familiar
with XML, but who have had little or no prior experience with
Prolog. You'll learn how document processing problems are addressed in
logic programming as compared to conventional procedural languages. A
programmer's introduction to Prolog will be followed by topics such
as:
* Strategies for parse tree representation
* Working with XML namespaces
* Reasoning about Documents: Syntax versus Semantics
* Dealing with overlapping hierarchies
Prolog is particularly useful for developing new XML processing
approaches, but it can be as efficient as other high-level languages,
even for routine transformations of XML.
My parents took me to the Semantic Web, and all I got was more markup
Some of the most celebrated XML-related developments in recent years
have focused on enriching markup languages with more expressive
semantic information: technologies like RDF, Topic Maps, and web
ontology languages all aim to bring knowledge representation to the
World Wide Web. So once you've got rich semantic information in a form
you can manipulate with XML processing tools -- what then? Prolog's
built-in inferencing capability provides a starting point for building
programs that reason with and about structured documents.
Extreme Markup Languages 2005
XML Processing in Prolog will be offered at Extreme Markup Languages
2005. The instructor will be David Dubin, a research scientist at the
University of Illinois at Urbana-Champaign. Extreme Markup Languages
is devoted to the theory and practice of markup languages from
industrial, academic, and other points of view. Further information
(including registration for the meeting and this tutorial) can be
found at
http://www.extrememarkup.com/extreme/
August 1, 8:30 AM-5:30 PM, Hotel Europa, Montreal, Canada
The XML of Programming Languages
The great appeal of XML for representing documents, business and
scientific data has its basis in a particular philosophy for
organizing work. As an XML user, you already understand the power of
describing things in terms of what they are, rather than how they look
or behave. You appreciate the flexibility gained by decoupling
definitions and descriptions from the details of how those data are
expected to be processed.
What would 'the XML of programming languages' be like? One answer,
of course, is that such a language would be an XML application itself,
capable of being processed just like (or even along with) other XML
markup.
This tutorial invites you to consider another answer: instead of a
programming language expressed *as* XML, we'll introduce one designed
around similar principles to those of declarative markup. Prolog was
developed at the Universities of Aix-Marseille and Edinburgh in the
early 1970s -- around the same time Charles Goldfarb was inventing
SGML at IBM. Colmerauer, Kowalski, and Roussel originally designed
Prolog to support research in natural language processing. It should
therefore come as no surprise that Prolog offers many benefits for
processing artificial languages, such as those expressed in XML.
Processing XML in Prolog
This tutorial will be an introduction to logic programming and to
problems, issues and strategies for XML processing in the Prolog
language. It is intended for experienced programmers who are familiar
with XML, but who have had little or no prior experience with
Prolog. You'll learn how document processing problems are addressed in
logic programming as compared to conventional procedural languages. A
programmer's introduction to Prolog will be followed by topics such
as:
* Strategies for parse tree representation
* Working with XML namespaces
* Reasoning about Documents: Syntax versus Semantics
* Dealing with overlapping hierarchies
Prolog is particularly useful for developing new XML processing
approaches, but it can be as efficient as other high-level languages,
even for routine transformations of XML.
My parents took me to the Semantic Web, and all I got was more markup
Some of the most celebrated XML-related developments in recent years
have focused on enriching markup languages with more expressive
semantic information: technologies like RDF, Topic Maps, and web
ontology languages all aim to bring knowledge representation to the
World Wide Web. So once you've got rich semantic information in a form
you can manipulate with XML processing tools -- what then? Prolog's
built-in inferencing capability provides a starting point for building
programs that reason with and about structured documents.
Extreme Markup Languages 2005
XML Processing in Prolog will be offered at Extreme Markup Languages
2005. The instructor will be David Dubin, a research scientist at the
University of Illinois at Urbana-Champaign. Extreme Markup Languages
is devoted to the theory and practice of markup languages from
industrial, academic, and other points of view. Further information
(including registration for the meeting and this tutorial) can be
found at
http://www.extrememarkup.com/extreme/