|
TECHNICAL
TRACK
TUESDAY, FEBRUARY 29
8:30 am
Morning
Keynote: Grass-roots XML
Jon Bosak, Sun Microsystems, conference co-chair
Jon Bosak of Sun Microsystems organized and led the
working group that created XML, and he continues to
oversee the development of the XML family of standards
as Chair of the W3C XML Coordination Group. He is
a founding member of OASIS and was a primary force
behind the development of the DocBook standard for
Unix and Linux documentation.
Session #1: Documents
9:00 am
Publishing
with XML and the Open eBook Specification
David
Goldstein, Chief Technical Officer, Versaware, Inc.
davidg@versaware.com
Biography:
David
Goldstein is the CTO of Versaware, Inc., one of the
companies pioneering the electronic publication industry.
Mr. Goldstein is active in XML based consulting and
serves on the authoring committee of the Open eBook
Specification. He also serves on both the executive
and technical committees of the EBX group working
on digital rights management for eBooks. Prior to
his work at Versaware, he was a Product Manager and
the Technical Evangelist of Accent Software, Inc.,
which produced some of the first multi-lingual Unicode-based
word processors and translation tools.
Abstract:
Sparked
by the growth in WEB access and the availability of
low cost, hand-held reading devices, the electronic
publication industry is on the verge of redefining
our relationship with the "printed" word. We are witnessing
a revolution in the way we read, access, search and
purchase "books". However, there are several obstacles
to overcome. These include the plethora of word processing
and publishing formats, the interoperability gap separating
traditional and electronic publishing, the representation
and encoding of a myriad of languages and alphabets,
and the high cost of re-processing content for various
target media.
The
Open eBook (OEB) Specification is part of an industry-sponsored
initiative to resolve many of these issues. The specification
seeks to have an immediate and direct impact on the
creation, advancement, and growth of a flourishing
eBook industry. Fully XML and Unicode compliant, this
specification supports feature-rich and media-enhanced
electronic publication. Content providers can now
feel confident that their documents can be delivered
on a wide range of reading devices without the need
to reprocess. Any OEB-compliant reading system will
display these publications in a manner that most faithfully
represents the original content, even in consideration
of limitations of a particular reading device.
This
presentation will introduce the goals and design
principles of the Open eBook Specification and will
detail its features, limitations and relationship
to other industry standards. Current tools for producing
and displaying OEB publications will be demonstrated
and future directions of the OEB authoring committee's
work will be described.
9:30 am
Ęsop:
A Browser for XML Documents and Open eBook Publications
Christopher Maden, Solutions
Architect, Yomu
crism@exemplary.net
Deborah Hooker, Vice President of Engineering, Yomu
deb@exemplary.net
Biographies:
Chris
Maden received his Sc.B. in electrical engineering
from Brown University; he worked for Electronic
Book Technologies for three years before taking
a position with O'Reilly & Associates, where
he developed a DSSSL-based print and on-line publishing
system. He contributed to the development of HTML
2.0 and XML, as well as the DynaText and DynaWeb
products at EBT, and is on the XSL Working Group.
He now provides OEB solutions to publishers and
works with the Ęsop architecture team at Yomu.
Deb Hooker has fifteen years of experience architecting
and building software solutions in domains as diverse
as astronomy, satellite remote sensing, telecommunications
and finance. She is the Vice President of Engineering
for Yomu.
Abstract:
Yomu
plans to become the world's leading distributor
of electronic books (ebooks). By ebooks, we mean
any large documents that are not suited to HTML
and Web-based viewing. This may include books previously
published in paper form, new texts written specifically
for this medium, internal corporate documents, or
any other large, complex document.
Towards
this end, we are developing sop, a high-quality
ebook browser, written in Java. Ęsop incorporates
technology to be a general-purpose XML browser,
initially focused on the subset required by the
OEB Publication Structure specification published
by the Open eBook Consortium (URL:http://www.openebook.org/).
We intend to provide high-quality cross-platform
rendering, addressing desktop and handheld systems.
The version of Ęsop seen at the conference will
be the first publicly seen version of our product.
We intend to have a stable release by June of 2000.
Ęsop
will enable the reader to enjoy a product that
takes into account human understanding of how
reading works. To this end, Ęsop incorporates
features that allow for accessibility, internationalization,
as well as less commonly understood features to
improve reading comprehension. We intend to demonstrate
some of these features, and clearly delineate
the advantages involved.
We
will demonstrate the rendering features of Ęsop,
including what we're calling "multi-view browsing".
Multi-view browsing allows the same rendered version
of a document to be flowed into different contexts;
for example, a user who is more comfortable with
traditional print books would prefer the illusion
of on-screen pages, while a more experienced computer
user may want a single continuous scroll. We will
generate three views of Ęsop during this presentation:
single page, two pages, and the scrolling view.
We
will discuss the processing model that Ęsop uses.
XML's primary strength, the separation of content
and behavior, leads to a question that the standards
have yet to answer: what happens when content
and behavior must be unified? The order of processing
a document, a stylesheet, and associated links
can produce different results. In addition, the
processing of "hints" in the instance (such as
XLinks, HTML-style tags, SVG, etc.) is unclear,
especially when coupled with stylesheets that
may reference these behavior-specific tags a little,
a lot, or not at all. We will share our thoughts
that led to our current prcoessing model, which
we believe produces the best results.
User
annotations are a key feature of an ebook browser.
Annotations can be shared among users of a book
without depending on a particular rendered view
of a book. A problem with document transformation
is that it can cause information based on the
reliability of a document's pre-transformed structure
to become unreliable. Tracking annotation locations
back to their origin in the source document and
reliably re-locating them in a new rendered view
is an interesting problem, and we will share our
thoughts on the various approaches we have considered,
where we arrived, and why.
We
have also developed a stylesheet object model
capable of representing the information contained
in both CSS and XSL stylesheets. Considerations
of performance and completeness that went into
creating the model will be covered, as will our
experiences implementing XSLT and XSL formatting
objects.
10:00 am
Implementing
the DOM for MathML in the IBM techexplorer
Hypermedia Browser
Sam Dooley, Staff Programmer, IBM Research
dooley@watson.ibm.com
Biography:
Sam
Dooley is a member of the Advanced Internet Publishing
Group at IBM Research, and is responsible for the
current implementations of several XML technologies
for the techexplorer Hypermedia Browser, including
DOM Level 1 interfaces for MathML and LaTeX node
types, parsing and rendering MathML presentation
and content elements, core XML language support,
and content markup extensions to the LaTeX language
used in techexplorer's interactive courseware offerings.
Research interests include applications of interactive
mathematical documents for symbolic computation,
scientific courseware, and the representation of
mathematical inference.
Abstract:
The
Document Object Model (DOM) allows an application
to expose a standard collection of interfaces so
that external programs, scripts, and applets can
modify the internal structure of a document. MathML
provides standard vocabularies for the communication
of mathematical information contained in internet
documents. Together they allow web content developers
to create applications with an unprecedented level
of sophistication in the treatment of the interaction
between a user and the mathematical content of a
document. These applications include interactive
scientific and technical documents, interactive
courseware, and mathematical software interfaces
that begin to realize the promise of XML technologies
for providing extensible documents with rich domain-specific
semantic content.
These
technologies have been brought together in the implementation
of the DOM interfaces for the MathML node types
supported by the IBM techexplorer Hypermedia Browser.
IBM techexplorer is a browser plugin for rendering
mathematical documents written in LaTeX and MathML,
and stands as the first open-standards platform
for developing interactive mathematical applications
for the internet. By using widely-supported markup
languages for mathematics (MathML, LaTeX) and industry
standard interfaces (DOM) for interacting with the
document structure, C++ and Java application developers
have open access to the mathematical content of
a document, and a high degree of control over its
presentation to the user.
IBM
techexplorer implements full support for the DOM
Level 1 interfaces for both the MathML 1.0 presentation
and content tag sets, and for a collection of
element types corresponding to commonly used TeX
and LaTeX syntactic structures. The implementation
of these interfaces involved adding DOM support
to a set of existing C++ classes for representing
and rendering documents written in LaTeX and MathML.
Several issues were addressed during the design
of the DOM implementation for techexplorer, including:
providing a set of application-independent DOM
implementation classes for C++, using delegation
instead of multiple inheritance, handling application-specific
treatment of the representation of text through
an application-independent DOM interface (Text),
exposing application-specific representations
of attributes through a uniform Attribute interface,
and the use of namespaces for handling collisions
between separate element vocabularies. Solutions
to these issues point to implications and design
requirements for the use of the DOM in other domain-specific
areas.
Session #2: News and Information
11:00 am
XML
Standards for News: The View at Halftime
Deren Hansen, Director, News Technologies & XML
Evangelist, WAVO Corporation
dhansen@wavo.com
Biography:
Deren
Hansen, WAVO's designated XML Evangelist, represents
his company in various XML industry and standards
activities. He serves as editor for the IDEAlliance's
PRISM (Publisher Requirements for Industry Standard
Metadata) initiative, and sits on the International
Press Telecommunication Council's IPTC-2000 working
group. An expert in the electronic news and information
industry, Mr. Hansen created WAVO's content acquisition
system and was one of the architects of WAVO's
NewsPak service. Mr. Hansen holds degrees from
the Massachusetts Institute of Technology and
the University of Delaware.
Abstract:
Continuous
news feeds are a priority for many web sites.
If someone wanting to add a news feed were to
search for XML news standards they would find
a bewildering collection of acronyms - NML, NITF,
XMLNews, PRISM, IPTC-2000, NewsML - whose associated
specifications and proponents share suspicious
similarities and tantalizing differences. Deren
Hansen, whose responsibilities at WAVO Corp. have
thrust him squarely in the middle of such things,
explains the origins of, the relationships between,
and likely future of the various XML standards
for news.
11:30 am
An
XML-Centric Architecture for Immersive Sports Coverage
Jeffrey E. Sussna, Chief
Architect, Quokka Sports, Inc.
jeff.sussna@quokka.com
Biography:
Jeffrey
E. Sussna has 15 years of experience designing
and building advanced software products. He has
led the development of cutting edge interactive
television and Web applications and end-to-end
content management systems for major media companies
including Time Warner, News Corp., and the Chicago
Tribune. Jeff has held technical leadership positions
at Silicon Valley companies such as Apple and
Oracle, and Multimedia Gulch companies such as
Ikonic Interactive and Quokka Sports. Jeff is
currently Chief Architect at Quokka Sports, Inc.,
where he is responsible for defining distributed
systems and protocols for live production of online
sports coverage.
Abstract:
Quokka
Sports produces immersive coverage of world-class
sporting events for Web, wireless, and broadband
environments. This coverage blends web content,
video, and multiple live data feeds, letting users
control their own perspective on an event. Immersion
production relies on a high-volume, low-latency
production process that integrates multiple data
types, data formats, tools, organizations, and
delivery platforms. Quokka is using XML as the
basis for a data-centric backbone for its end-to-end
immersion production environment, the Quokka Sports
Platform (QSP).
QSP
uses XML for several fundamental types of production
data:
*
Metadata about individual assets (represented
using RDF)
*
Live data streams for such things as timing, position,
telemetry, and biometrics
*
Concept-based organizational and search structures
(also represented using RDF)
*
Platform-independent content publishing
*
Content packaging for distribution channels
*
Dynamic content serving
*
Client-side navigation
XML-based
production data enable interfaces between people
organizations, tools, and components throughout
the QSP lifecycle. These interfaces include:
*
Event venues and the Quokka production studio
*
Third-party content creators and the Quokka production
studio
*
Distributed Quokka production facilities
*
The Quokka production studio and content distribution
partners
*
Dynamic and static content serving systems
*
Event producers and end-users (via client-side
content navigation)
This
presentation describes XML's central role
in the design of a loosely coupled distributed
system that integrates multiple content types,
tools, and organizations. It begins by introducing
QSP and describing the various XML-based formats
used by QSP. It explains how these formats
help QSP meet its goals of openness, flexibility,
and adaptability. Finally, the presentation
discusses XML adoption issues, including design
decisions and tradeoffs, internal knowledge
transfer, and cross-organizational perceptual
issues.
Session #3: Technologies
2:00
pm
Introduction
to XUL, the XML-Based User interface Language: An XML
Application for Defining the User Interface of Software
Applications
Eric Krock, Senior Product Manager,
Netscape Communications
ekrock@netscape.com
Biography:
Eric
Krock is Senior Product Manager for Netscape Communicator
at Netscape Communications. He is responsible for
component technologies including the Netscape Gecko
layout engine and its support for W3C standards, JavaScript,
the Open JVM Interface, and the Mozilla Plug-in API.
Previously he worked as a Technology Evangelist responsible
for JavaScript, Dynamic HTML, and Commerce Applications.
Abstract:
Topics
covered in this session:
*
history of cross-platform GUI development: Windows,
Macintosh, Motif, and cross-platform solutions
*
maturation of web standards over 1994-2000
*
power offered by web standards today
*
HTML, XML, and XML Namespaces for representing structured
data
*
RDF for representing resources
*
CSS for formatting
*
DOM for positioning and event handling and its CSS
interface for style control
*
JavaScript as scripting language
*
motivation for XUL
*
XUL foundation concepts: packages, widgets, box
and spring layout
*
XUL markup
*
XUL examples: toolbar, pulldown menus, bookmarks
with RDF, dialogs, windows, trees
*
real-world example: UI of Mozilla browser
*
modifying the UI by editing XUL files; ChromeZone
themes on mozillazine.org
*
XUL futures: XUL overlays to enable modularity;
XUL dynamically loaded from server
*
resources for learning more
2:30 pm
Solaris
WBEM: Sun's Implementation of CIM Over HTTP
Ed Mooney, Staff Engineer,
XML Technology Center, Sun Microsystems, Inc.
Ed.Mooney@Sun.com
Biography:
Employed
by Sun since 1993, most recently on the engineering
staff in the XML Technology Center. Past assignments
have all been in the realm of electronic publishing,
including many dues paid converting Frame to SGML.
Chair of the Interoperability Working Group within
the Distributed Management Task Force.
Abstract:
Sun
Microsystems will deliver support for Web Based Enterprise
Management with Solaris(TM) 8 (Q1 2000)[1]. This technology
enables Solaris 8 applications to manage other WBEM-enabled
platforms, and other WBEM-enabled applications to
manage the Solaris Operating Environment. Java and
XML lie at the heart of this support.
WBEM
is an initiative of the Distributed Management Task
Force[2]. Of the major computer vendors among the
DMTF's membership, the following sit on its board
of directors:
*
Cisco
*
Compaq Computer Corp.
*
Dell Computer Corp.
*
Hewlett-Packard Company
*
IBM/Tivoli Systems, Inc.
*
Intel Corporation
*
Microsoft Corporation
*
NEC Corporation
*
Novell
*
Santa Cruz Operation, Inc. (SCO)
*
Sun Microsystems, Inc.
*
Symantec Corporation
XML
forms one foot of the tripod on which WBEM rests.
The Common Information Model (CIM) and a mapping
of CIM to HTTP form the other two feet.
The
Common Information Model (CIM) is an object-oriented,
language-neutral data model. It comprises a specification[3]
and schema[4].
The
CIM specification defines a metaschema. Classes
and instances occupy the metaschema along with other
abstractions such as qualifiers, properties, methods
and associations.
The
schema describe the components to be managed. Schema
are written in plain text using a language called
Managed Object Format (MOF).
The
CIM specification recognizes three types of schema.
The core schema provide a small set of classes,
associations and properties for analyzing and describing
managed systems. The common schema provide definitions
for technology-independent systems, applications,
networks and devices. Extension schemas are technology-dependent
extensions to the common model. The DMTF provides
the core and common schema. Each vendor provides
its own extension schema.
The
"Specification for the Representation of CIM in
XML"[5] provides a syntax both for declaring managed
objects in XML and for XML encodings of operations
against a WBEM-enabled device. The "Specification
for CIM Operations over HTTP"[6] maps a set of 23
operations onto HTTP.
WBEM
services will make the Solaris Operating Environment
manageable by tools from enterprise vendors other
than Sun. These services include 100% Pure Java(TM)
implementations of:
*
A manager for CIM objects
*
A MOF compiler
*
An API for writing object manager clients
The
interfaces support two transports: Remote Method
Invocation (RMI) and HTTP. The HTTP transport
implements the "Specification for CIM Operations
over HTTP." It was built using Project X, Sun's
Java implementation of DOM and SAX, and the Java
Servlet Development Kit, also from Sun.
The
HTTP transport exposes no public interfaces. The
goal is for any functionality accessible by way
of RMI also to be accessible by way of HTTP. However,
the HTTP transport awaits key technologies from
the DMTF to be as functional as RMI. Among these
are models for events and security.
Many
of the non-public interfaces of the HTTP transport
are devoted to mapping Sun's WBEM client API to
the functional profiles specified in Section 2.6
of "Specification for CIM Operations over HTTP."
The design favors DOM over SAX. Even so, it doesn't
use DOM methods directly. In general, it parses
strings into DOM at the latest possible opportunity.
It uses Project X's TreeWalker class to parse
DOM objects into CIM objects.
We
have no hard performance statistics. Subjectively,
performance over HTTP is noticeably slower compared
to RMI, but acceptably responsive from a web-surfer's
perspective.
Though
perhaps a toddler today, WBEM technology is growing
fast. The DMTF published the standards only after
Sun, Microsoft, and Hewlett-Packard had shown
that their prototypes could interoperate. It is
hard at work on models for events, queries, security
and discovery which it also plans to realize in
XML.
1.
http://www.sun.com/solaris/wbem/
2.
http://dmtf.org/
3.
http://dmtf.org/spec/cim_spec_v22/
4.
http://dmtf.org/spec/cim_schema_v22.html
5.
http://dmtf.org/download/spec/xmls/CIM_XML_Mapping20.
htm
6.
http://dmtf.org/download/spec/xmls/CIM_HTTP_Mapping10.
htm
3:00 pm
What
XML Schema Designers Need to Know About Measurement
Units
Frank Olken, Computer Scientist , Lawrence Berkeley
National Laboratory
olken@lbl.gov
John McCarthy, Computer Scientist, Lawrence Berkeley
National Laboratory
jlmccarthy@lbl.gov
Biography:
Frank
Olken is a database researcher at Lawrence Berkeley
National Laboratory. He currently works on metadata
and related standards (ISO 11179, XML Schema, XML
Query Language) and various informatic issues (data
exchange, distributed systems, etc.) related to electric
power systems. His other interests include statistical
data management, sampling from database, OLAP, genomic
databases, workflow management, ....
John McCarthy is a database administrator and researcher
at Lawrence Berkeley National Laboratory. He currently
also works on metadata standards (ISO 11179, XML Schema
Language).
Abstract:
On
September 23, 1999, NASA lost the Mars Climate Observer
spacecraft (at a cost of $125 Million). "The 'root
cause' of the loss of the spacecraft was the failed
translation of English units into metric units in
a segment of ground-based, navigation-related mission
software, as NASA has previously announced," said
Arthur Stephenson, Chairman of the Mars Climate Orbiter
Mission Failure Investigation Board.
This
loss reminds us of the importance of measurement units
for data exchange in various fields (navigation, engineering,
architecture, medicine, science, and commerce). Misunderstandings
of measurement units are not simply expensive - in
medical settings and aircraft navigation they can
be fatal. As the Failure Investigation Board noted,
NASA contractors failed to fully specify the interface
(i.e., measurement units) between the two software
packages. While NASA contemplates various management
review mechanisms to prevent recurrence of such problems,
we believe the appropriate answer is to effectively
extend the type systems used to specify data interchanges
(e.g., XML Schemas for XML documents) to include measurement
unit information. Hence, it would be possible to automatically
detect (and often to automatically resolve via conversion)
inconsistent measurement units, much as we now routinely
detect the need to convert among integer, single precision
floating point and double precision floating point
numbers in computer programs.
We
review the classical semantic theory of physical
measurement units known as dimensional analysis.
This theory was developed over the past 100 years
by physicists, chemists and mathematicians. Each
kind of measurement has a "dimensionality", e.g.,
mass, length, time, length/time (i.e., speed), etc.
The dimensionality may be specified as the ratio
of two monomials in which the variables (indeterminants)
represent the basis dimensions e.g. in the SI system
of units, mass, length, time, current, .... Thus
dimensionality for speed would be (Mass**1/Length**1).
The dimensionality for energy is (Mass**1 * Length**2
/ Time**2 ). Note that dimensionality is a partial
specification of the semantics of a measurement.
Also note that given a specified list of basis dimensions,
we can specify the dimensionality of a measurement
as 2 exponent vectors, where we have restricted
the exponents to be non-negative.
In
contrast measurement units are representational
aspect of a measurement. Typically they are specified
as a scale factor with respect to a set of basis
units (one basis unit for each basis dimension).
Thus 1 inch is defined as 2.54 * 10**-2 meters.
The
reason for using a ratio of two monomials for dimensionality
is that this permits us to distinguish "dimensionless"
quantities, such as mass ratios (mass/mass), molar
ratios (moles/moles), or volume ratios (length**3/length**3).
Such a requirement arises with respect to dimensionless
(ratiometric) specification of concentrations.
Most
common dimensionally consistent unit conversions
can be accomplished by dividing by the ratio of
the unit scale factors with respect to a common
set of basis units. However, dimensionally inconsistent
conversions, e.g., mass to volume are quite common
in commercial applications. Such conversions are
more complex, relying on knowledge of materials
properties (such as density) which vary with the
condition (e.g., temperature) of the material.
Not
all unit conversions are simple multiplications.
Temperature conversions are affine transformations,
accounting for differing origins of the temperature
scales. Also note that "temperature coordinates"
and "temperature intervals" are different sorts
of measurements and have differing unit conversions.
Lamentably, this distinction is not always made
explicitly. Similar distinctions must be made between
spatial interval lengths and spatial coordinates
as well as temporal intervals and coordinates. Temporal
coordinates are commonly referred to as dates or
datetimestamps.
Having
reviewed this generic discussion of dimensionality
and measurement unit conversion we proceed to discuss
how measurement units (and dimensionality) could
be systematically encoded in XML (either in schemas
or documents). We observe that usage of conventional
abbreviations (miles/hour) would require the development
of an application specific parser. Instead we suggest
a systematic mapping of conventional units designations
into legal XML names (miles_Per_hour). Such names
could be used to reference definitions (dimensionality
monomials, scale factors, ...) via designated namespaces
(e.g., ISOMeasurementUnits). Such an approach, while
more verbose than conventional abbreviations, would
be simpler to implement and maintain. We provide
detailed examples of how this could be done.
We
suggest that such measurement unit encodings should
be standardized for use in conjunction with the
W3C XML Schema Languages specification. We envision
that such measurement units specifications would
typically be specified in the XML schema for a data
exchange document as an extension of the datatype
specification. We consider the implications of this
with respect to the extensibility and annotation
mechanisms proposed for the XML Schema Language.
Finally,
we discuss the relation of this work to previous
efforts concerning measurement units and dimensionality
specification in the standards, database, and
programming language communities.
Session #4: Databases
4:00 pm
RDBMS
in XML, XML in RDBMS: Architectural Considerations
Dale Hunscher, CEO, South Wind
Design, Inc.
dale@swdi.com
Biography:
Mr.
Hunscher has worked as a programmer, systems analyst/designer/architect,
engineering manager, and executive in the software
field since 1983. His company, South Wind Design,
Inc., has provided OEM software development and
engineering services since 1993.
South
Wind originally specialized in ODBC driver and
application tool development, and expanded beginning
in 1996 into technical support tools and Internet
development, especially Web-enabled database applications.
For the past eighteen months Mr. Hunscher has
played a central role in the development and deployment
of an XML transaction processing system that applies
30,000 - 100,000 XML documents per day against
a 100+ GB Oracle database.
Abstract:
As
XML becomes a key enabling technology for electronic
commerce (e-commerce) and electronic data interchange
(EDI), interplay with enterprise relational data
base management systems (RDBMS) is inevitable.
How will this interplay be structured? Will the
RDBMS paradigm drive document structure, or will
XML document structure be mapped onto RDBMS structures?
Or will there be synergy between the two?
This
presentation starts with a review of the architecture
and terminology of RDBMS entities and relationships
and XML document structures. While many similarities
exist in the two terminologies, do the terms
held in common really mean the same thing? After
examining this question in some detail, we identify
the key differences between the two conceptual
frameworks. With this as our foundation we can
begin to look at some possible mappings between
RDBMS and XML, and evaluate the fit and utility
of each.
XML
documents are usually transactional in nature,
representing an event that will cause a change
in state in the systems on each side of the
transaction. For example, an online purchase
triggers the simultaneous transfer of goods
from seller to buyer and cash equivalent from
buyer to seller, mediated by one or more financial
institutions that handle the accounts of the
buyer and the seller, as well as other aspects
of the transaction, such as the financial network
over which the transaction is transmitted. The
transaction causes changes in the accounting
mechanisms on all sides of the transaction.
The transaction means different things to each
of the involved parties. For one it may denote
the movement of value from one asset account
to another, and for another, the reduction or
increase in both a liability and an asset account.
An
XML document transaction set can easily represent
the details of these financial conversations
for all the transaction processing systems along
the way. How are such transaction documents
best mapped to RDBMS transactions? One approach
is to use XSL to transform a transaction into
one or more SQL statements. We'll look in some
detail at a simple example of how this can be
done, to expose and analyze some of the pitfalls
involved.
Finally
we'll consider the challenge of representing
RDBMS data sets (SQL query result sets) as XML
documents. There are a number of design aspects
to consider in mapping the data set to a document:
*
The SQL result set is tabular in structure.
XML documents are inherently hierarchical. What
is the best way to map from one to another?
*
Because one of the great strengths of RDBMS
is query flexibility, SQL result set meta-data
is highly dynamic. For best results an XML document
should have a fixed structure, whether in the
form of a Document Type Definition or an XML-Data
schema. What is the best way to manage the flexibility
of the SQL result set in producing XML document
structures?
*
XML allows the document designer two mechanisms
for storing instance data: attribute values
and element content. Which are most appropriate
for representing result set content?
*
A result set can contain an arbitrary number
of rows. Should each row be represented as an
XML document, or should a single document "container"
represent the entire result set? Moreover, most
RDBMS APIs support block cursors, which provide
for the retrieval of multiple-row subsets of
the result set. What is the best way to handle
block cursors in mapping to XML documents?
Along
the way we'll learn about some of the RDBMS vendor
approaches to handling interactions with XML documents
and take a look at WDDX, a proposal for mapping
RDBMS result sets to XML documents.
4:30 pm
Enabling
Your Business Data for XML with DB2 XML Extender
Jane Xu, Senior Software Engineer, Member of DataBase
Technology Institute, Software Group, IBM Corporation
jxu@us.ibm.com
Josephine Cheng, Distinguish Engineer Manager of DataBase
Technology Institute, Software Group, IBM Corporation
Biography:
Jane
Xu is a senior software engineer at IBM Database
Technology Institute. She is currently the architect
and technical lead of DB2 UDB XML Extender. Jane
Xu received her Ph.D in Computer Science from
the University of Southern California in 1990
and since then has been worked for IBM. Prior
to that, she obtained her B.S. and M.S. degree
in Computer Science from University Hawaii, Manoa.
She worked on architecture and strategy of distributed
file and storage systems while she was with IBM
Storage Systems Division. After joining IBM Software
Group, she was the technical lead of IBM Digital
Library multimedia support and Net.Data products.Jane
has published many publications in technical magazines
and conferences. She filed certain number of patents
for IBM.
Josephine Cheng is an IBM Distinguished Engineer
at Santa Teresa Laboratory and a member at IBM
Academy of Technology.She is also a manager at
the Database Technology Institute responsible
for advanced technologies for IBM database products
including the product development of DB2 WWW Connection,
Net.Data, XML Extender and DB2 Everywhere. Josephine
received YWCA's 1996 Tribute to Women and Industry
Award. She is a co-editor and co-author of "Web
Gateway Tools" published by Wiley Computer Publishing,
4/97. She serves in National Research Council
Information Panel for technology assessment since
1998. Josephine received the B.S. degree in Mathematics
and Computer Sciences and the M.S. in Computer
Sciences from the University of California, Los
Angeles.
Abstract:
E-business
requires stability, scalability and security.
When e-business is adopting XML, DB2 allows you
to leverage the power of XML to your existing
business data, where XML can be your Web document
content or your B2B data interchange format. DB2
XML Extender is a new extender of the DB2 extender
family. It is a SDK which provides XML support
for DB2. It not only makes DB2 as the repository
of XMLdocuments and DTDs, but also enables existing
relational data for XML. With XML Extender, you
can decompose XML data into relational tables,
and generate XML documents from data stored in
DB2.
When
companies turn to XML for their Web content,
a repository for XML documents is needed. XML
Extender provides new data types that let you
store XML documents in DB2 databases and new
functions that assist you in working with these
structured documents, including fast indexing
on XML elements and attributes. Entire XML documents
can be stored in DB2 database as character data
up to 2GB each, or stored as external files
but still managed by DB2. Retrieve functions
allow you to retrieve either the entire XML
document or individual elements and attributes.
XML Extender provides ways to ensure high performance
queries.Ż You can specify elements or attributes
in the XML document to be converted into standard
SQL data types, and to be indexed for facilitating
the high performance search. XML Extender also
works the DB2 Text Extender, which has a new
section search feature that supports the structured
text search for XML.
While
electronic data interchange(EDI) has been around
for many years, implementing EDI systems to
communicate with supplies and trading partners
among companies is still expensive. XML-based
data interchange formats are opening up B2B
communication to all companies with lower cost.
Because of the flexibility and openness, XML
is becoming the standard to define data interchange
formats. DB2 XML Extender allows you to transform
XML content to a collection of traditional data
types in new or existing DB2 tables. It provides
stored procedures that decompose the XML content,
convert elements and attributes into SQL data
types and stores them in relational tables.
Stored procedures are also provided to compose
XML documents from your existing business data
in DB2.
The
mapping between relational tables and XML documents
via the Data Access Definition(DAD), an XML
formatted specification. In DAD, you can specify
how to extract XML elements and attributes from
entirely stored XML documents to create indexing.
You can specify the way to compose or decompose
XML documents from or into relational tables
via the explicit or implicit SQL query,Ż based
on the XSLT data model. The mapping is directly
from relational tables to desired XML structure
without further intermediate transformation.
This gives XML Extender better performance.
DB2
XML Extender is also the repository of DTDs.
With DTDs stored in DB2 XML Extender provides
validation feature to validate entire XML documents
stored as DB2 column data, as well as to validate
input or generated XML documents for decomposition
and composition.
The
DB2 XML Extender comes with a visual administration
tool is provided to easily define the mapping
of elements and attributes from and XML document
into columns and tables.
In this presentation, we will discuss the
following topics:
*
How to determine XML data storage and access
method.
*
How to stored entire XML documents in DB2
and create index on elements and attributes
for fast search.
*
How to generate XML documents from data in
relational tables, and stored un-tagged data
from XML documents into traditional SQL data
columns.
*
How to validate XML documents which XML documents
stored in the same column with more than DTDs.
5:00 pm
Building
Oracle8i XML Applications with XML, XSLT, XSQL
Pages,
and Java
Steve Muench, Lead XML Evangelist, Oracle Corporation
smuench@oracle.com
Biography:
Steve
Muench is Consulting Product Manager on the Business
Components for Java Development Team, and Oracle's
lead technical evangelist for XML. In his ten
years at Oracle, he's been involved in the support,
development, and evangelism of Oracle's application
development tools and database. In addition to
being Oracle's primary representative to the XSL
Working Group at the W3C, he's been a driving
force in helping Oracle development teams from
the database server, to application server, to
tools, to packaged applications weave XML sensibly
into their future development plans. He's busy
working on a book on using Oracle8i, XML and XSLT
for O'Reilly, due out sometime in the year 2000.
Steve has presented at the following industry
conferences: XML98 (keynote), XMLEurope99 (Keynote),
XML99 ("Relational Databases to XML"), and two
XMLOne conferences, in addition to JavaOne 1999
and countless Oracle OpenWorld conferences.
Abstract:
Oracle's
XSQL Servlet is a free yet amazingly powerful
utility that can be used with any Servlet Engine
to more easily build applications that combine
the power of SQL, XML, and XSLT without programming.
In
the presentation we walk through numerous demonstrations
of applications you can build using XSQL Pages
by combining your knowledge of SQL with XSLT Transformations
to make magic happen.
We'll
explore all of XSQL's built-in action elements
and step through the details of building your
own custom actions to implement virtually
any kind of server-side XML processing your
particular application might need.
MANAGEMENT
TRACK --
TECHNICAL
TRACK ( March
1, March
2)
|
