PRE-CONFERENCE TUTORIALS • 3 November 2008

A Business Approach to Metadata Deployment
Jonathan Geiger, Intelligent Solutions

Metadata is often viewed as a technical issue, and that is due in part to its name. Managing metadata is really about managing the information about the enterprise asset of “data” – an asset for which Information Technology is ultimately responsible. The business view of metadata is critical for positioning metadata in the context needed to gain executive and business support and to realize maximum value from metadata. It is generally recognized that metadata is an essential component of any successful BI effort, and that metadata management holds great potential value for all IT initiatives. In addition, metadata can help companies address regulatory and data quality issues.

The increased familiarity with metadata has not been accompanied by an increase in practical, cost-efficient solutions available to help manage metadata. Metadata management efforts today still tend to be lengthy and expensive propositions, which require a great deal of patience and perseverance before tangible results are seen. Some software vendors are including rudimentary metadata management functionality in their tools, but these are only point solutions. Other vendors offer enterprise repository solutions for metadata management, but these have proven to be very expensive and difficult to implement. This tutorial will provide an alternative strategy to the implementation of a centralized metadata repository. It will describe how an organization can put together a very practical metadata strategy that leverages the tools and investments it already has, and how it can apply traditional application development techniques to create its metadata management environment.

• Data as an enterprise asset
  • Business Perspective
  • Major Roles and Responsibilities
• Metadata Overview
  • Traditional View
  • Recommended View
  • Types of Metadata
  • Role in Business Intelligence
  • Role in Quality Management
• Metadata Deployment
  • Architectural Options
  • Data Model and Database (a.k.a. Metamodel and Repository)
  • Development Methodology
  • Business Application Example
• Metadata Strategy Components
• Metadata Technology
• Action Plans

ABCs of Information Quality
Larry P English, INFORMATION IMPACT International

While the high—and mostly hidden—costs of poor quality information hurts both competitiveness and profits, IQ problems cannot be solved without understanding and applying sound quality management principles to information as a product of our business processes. Information quality management is not an academic exercise—it is a combination of quality principles, processes and culture transformation required for business performance excellence in the emerging, realized Information Age.

World-class organizations apply the same quality principles, such as Deming’s Fourteen Points, Kaizen, Quality Function Deployment (QFD) and the Baldrige Criteria for Business Performance Excellence to information. This presentation addresses how these principles and techniques apply directly to information as a product and knowledge workers as information producers.

In this tutorial Mr. English describes the fundamental principles of information quality. He describes how an organization can improve the quality and value of its information resources. He describes metrics for measuring information quality and the management principles for implementing an effective information quality environment. Mr. English describes how organizations have successfully implemented information quality processes to improve the effectiveness of their business and information system processes.

   A. Assessment: Introduction and Processes of IQ Appraisal
   B Betterment: Applying the Process Improvement Cycle to Information Processes
   C. Culture Transformation: Creating a Sustainable Environment for Quality Information

Global Data Excellence Framework (GDE-F WeA-9.0) from Vision to Value Generation
Walid el Abed, Global Data Excellence

During this tutorial we will walk you in an interactive way through a comprehensive methodology with practical examples on how to define and execute a Global Data Excellence Framework from vision to Business value generation in global enterprises:

• Setting the vision and the mission of data quality aligned with the business objectives.
• Elaborating on the business value drivers.
• Establishing the Framework to manage Data Quality from organizational, systems, and process perspective including the Continuous Excellence Process and the Data Quality Dimensions.
• How to measure data quality in a business driven way based on business rules.
• Data Quality KPIs model
• The business rules role and a data model to manage business rules.
• Data Governance model and its execution in a global environment.

The GDE-F WeA-9.0 has proven to be successfully implemented in different global enterprises thanks to its practical and pragmatic approach to data quality.

Using Agile Best Practices in Business Intelligence Projects
Sander Hoogendoorn, Capgemini
Sandra Wennemers, Capgemini

Due to their highly dynamic and evolving character, business intelligence projects can benefit greatly from applying short iterative development cycles and frequent customer feedback, as is custom to agile software development. Applying an agile approach to business intelligence projects, such as Scrum, XP or Smart, is most natural.

During this interactive session Sander Hoogendoorn and Sandra Wennemers will demonstrate how agile software development matches data warehouse development projects and show how UML modelling techniques can be used in planning and designing such projects. In particular, a requirements technique named smart use cases is used to define not only reporting and analysis, but also in ETL, covering up to 80% in data warehouse projects.

• What is agile software development? What are its best practices?
• How can agile software development help me improve my data warehouse projects?
• What are smart use cases and how do I model smart use cases?
• How do smart use cases fit data warehousing?
• How do I estimate and plan my data warehouse projects using smart use cases?

What Next for Business Intelligence - Changing Gears or Shifting Paradigms?
Barry Devlin, 9sight Consulting

Business Intelligence as an independent discipline is dying. Much of its former value has migrated to the operational environment as day-to-day decision-making is automated for short-term gain. Meanwhile, longer term decisions stand strangely disconnected from the underlying reality of business, market and organization. It’s clear that the days of the stand-alone data warehouse are numbered; but what next?

It’s decision-time for BI itself. We must move to the next level in the value chain for decision-making—engaging the strategic decision makers and enabling them to integrate in the business as a whole. Over the coming few years, BI managers will have to focus simultaneously on two goals: incorporating the old stand-alone BI infrastructure into the mainstream IT environment and creating a new future “business insight” environment.

We must begin to envisage a new type of “Highly Evolved Business”. This will require integration across all dimensions of its activities, as already being driven by SOA and Web 2.0 collaborative computing. And BI must become a driver and integral part of this new world.

This tutorial describes the implications of current technological changes in the IT world and describes at a high level a possible future architecture and how we could begin to implement it, covering:

• A review of the old paradigms and emerging trends in BI and where they deliver business value
• The use of SOA and Web 2.0 to enable pervasive business and people-driven processes
• Moving from a traditional data warehousing architecture to a wider, cross-business and cross-functional view
• Breaking down old barriers between the different parts of the IT and business organisations and engaging the corporation in the next wave of information strategy

Effective EIM and Data Governance for Business Leaders
John Ladley, IMCue Solutions

EIM is a business issue. If information is truly an asset, then the entire business must be engaged. Business leaders are called upon to "do governance", learn to be stewards, data owners, and change agents while still accomplishing their day-to-day responsibilities. Research has shown there to be four critical success factors to ensuring data governance programs are executed successfully. None of these have anything to do with IT, software, or measurements of IT. This tutorial will explain how to manage these factors, and what role the business executive must play.

• The Business Case
• Leadership and Will
• Cultural Maturity
• Ability to Execute

Half Day
Tutorial
09:00-12:40

Data Modelling For Business Rules
Graham Witt, Ajilon

This workshop takes participants through a proven practical approach to modelling data for business rules. While some theory is covered, the workshop focuses on ensuring that the rules required by the business are documented and effectively implemented. Some interesting rules are explored such as temporal rules and rules constraining recursive relationships.
We will cover the following topics:

• The business perspective: rule motivation, rule enforcement, handling rule violations
• The analysis perspective: rules and data, rules and processes, rules and events
• Industry trends: structured business vocabularies, the SBVR (Semantics of Business Vocabulary and Business Rules)
• Discovering and verifying rules
• Implementing rules: data structures and properties, declared constraints, data content

From SBVR to Logical Data Models
Donald Chapin, Business Semantics
John Hall, Model Systems

Data’s value depends on how accurately it records the reality of business situations and how well it reflects what the people in a business think.
Until recently, both were entirely the responsibility of logical data modellers.

Now, however, “Semantics of Business Vocabulary and Business Rules” (SBVR) – a new OMG standard - enables business people to model their business, unambiguously and formally, using their own concepts and terms.

The resulting terminology-based model represents the real world of the business. For specification of information systems, it needs to be transformed to a logical data model. This tutorial is about how to make that transformation, including:

• SBVR, and how terminology-based business models are developed
• Making the transformation to a logical data model
• Traceability

Designing, Implementing and Maintaining SOA Solution and Data Services
Mehmet Orun, Genentech

This workshop intends to provide a deeper, end-to-end point of view based on actual trial-and-improvement in a corporate environment over the past few years. It will go beyond the marketing statements around SOA and focus on how it is different and how organizations can implement it. It will stay above the pure development focus around web services to describe how participants can plan and design for reuse. It will also describe how traditional data management skills are essential to SOA in defining canonical schemas, establishing data standards, and ensuring flow of better quality data across the service bus. The workshop will be broken into three main parts:

• Premise of SOA and how it differs from traditional N-Tier solutions
• How to adjust your SDLC Methodology for SOA
• Designing Data Services: Canonicals, Data Processing, and Services Flow

Data Modelling Challenges within EIM
Steve Hoberman, Steve Hoberman & Associates

This tutorial contains a completely new set of easy, moderate, and difficult data modeling scenarios that we will face within the field of Enterprise Information Management. After mastering easy challenges, you'll advance to more moderate and difficult challenges. This is not just a lecture. You'll get hands-on experience in areas such as:

• Gaining business agreement on enterprise definitions.
• Identify the analyst’s and modeler’s roles in implementation Enterprise Resource Planning (ERP) reporting solutions.
• Modeling unstructured data.

Successful Master Data Management
Malcolm Chisholm, President, Askget.com, Inc

Master Data Management (MDM) is increasingly seen as an imperative for many enterprises. Yet it is not a single project, is not targeted to a single user group, and rarely has a single sponsor. This tutorial examines the dimensions of an MDM programme, and a framework is presented for organizing the different activities involved. The MDM programme is placed within the context of an EIM function which is responsible for relevant aspects of data governance. Special attention is paid to issues in architecture, integration, and MDM services.

Attendees will learn:

• What is Master Data?
• Components of a Master Data Management programme
• The role of EIM and the elements of data governance
• Important technical Issues in MDM
• The role of metadata, including semantic management, and Business Rules Engine functionality.

Introduction to the Zachman Framework for EA
John Zachman, Zachman International

An introduction to the background, rationale and logic of the Zachman Framework for Enterprise Architecture. It will first explore some definitive reasons for the appearance of the Zachman Framework and will then provide an overview of the basic logic of the Framework itself which is derived from the precedent established in the older disciplines of Architecture and Construction, Engineering and Manufacturing. Mr. Zachman will show the importance of ensuring that long term fundamentals and building blocks are addressed and retained into the future. John is the originator of the "Framework for Enterprise Architecture" which has received broad acceptance around the world as an integrative framework, or "periodic table" of descriptive representations for Enterprises.

Using Unstructured Data To Enable A Better Business Decision
Krish Krishnan, Sixth Sense Advisers Inc.

Unstructured data is the newest entrant into the data warehouse. For 30 years plus, we have used the structured data available from our transactional systems to perform reporting and analysis. While this data has been very useful in solving a large business need, it did not answer the needs of the data mining and textual mining community. Over the years, we have tried to integrate unstructured data into the data warehouse but have not been successful. With the advent of DW 2.0 and a new series of techniques to integrate the textual data into the data warehouse, we now have the ability to provide a holistic view of the intelligence about the business. From this session you will learn:

• What is unstructured data?
• How to integrate unstructured data into the data warehouse
• Benefits of unstructured data