Vision and Strategy

1
Vision and Strategy

• Eugene Wong
• CEO, Versata, Inc.

2
Outline

• Review of Progress
• Technology and Product Vision
• Strategy

3
Progress

• Focus on Customer Success
• Major customer successes
• BT
• AMS
• Dept. of Labor (timely intervention)
• Signs of Reversal of Decline

4
Revenue History
8
6
M
software
4
2
Q1/01 Q2/01 Q3/01
Q4/01 Q1/02 Q2/02
Q3/02 Q4/02
Q1/03
5
Revenue Trend 2Q Moving Average
8
6
M
software
4
2
Q2/01 Q3/01 Q4/01
Q1/02 Q2/02 Q3/02
Q4/02 Q1/03
6
Revenue Trend 2Q Moving Average
8
6
M
software
4
2
service
Q2/01 Q3/01 Q4/01
Q1/02 Q2/02 Q3/02
Q4/02 Q1/03
7
The Software Inferno
paradise
Application
purgatory
Middleware
Systems
hell
8
The Software Inferno
paradise
Application
Versata
purgatory
Middleware
Systems
hell
Any resemblance to former or current employees is
purely coincidental
9
Middleware Vendors Lament

• Multiplicity of Platforms
• .net, J2EE, WAS, WLS, Linux, zOS
• Volatility of Change
• Immature technology
• Tyranny of Standardization
• Destroys differentiation
• Good tools get absorbed

10
The Escape - Versata Style

• Domain specific solution sale
• User management of business logic
• Need to understand what these terms mean

11
Solution Themes According to IBM

• e-business On Ramp (web sites)
• e-business Enablement (component integration)
• On Demand Applications (web services)
• Accelerating Development Life Cycle
• Full life cycle methodology on a common platform
• Enterprise Modernization
• End-to-end solution for reuse of existing assets

12
Business Logic

• As used, the term is marketing terminology
• It is anything but logic
• Logic a formal system of reasoning
• Gift from the Greek (Aristotle)
• Appropriate for our context
• Business Logic (my proposed definition)
• End to end representation of complete application
  lifecycle (according to some modeling paradigm)

13
Application Life Cycle Principal Stages
Conceive
clarify
Define
prototype
Design
Many participants Many stages At different
times Of long duration
validate
Upgrade
Implement
test
tune
Deploy
Maintain
14
Need a Common Universe of Discourse

• Conceptual Framework
• Vehicle for clear thinking
• Medium of Communication
• Among all participants at all stages
• Complete record
• of all relevant information
• In structured form
• Executable specification
• In short End-to-end MDA (Model Driven
  Architecture)

15
Concepts important for the modeling paradigm

• Layered and nested
• Successive and piecewise refinement (iterated
  decomposition)
• Self-similar
• No new types as one drills down
• Primitive
• Portion of application that can be executed
  without further decomposition
• (low level small portion, high level large
  portion)

16
Proposed Model

• Database Schema Process ( less important
  pieces)
• Each part consists of a nested set of directed
  graphs
• Each graph has only one type of nodes and one
  type of branches
• Semantics are captured in the content of the
  nodes
• Process node activity
• Schema node object class

17
Example Purchasing Concert Ticket
process
Buy tickets
Choose seats
Select event
schema
Events
Available Seats
18
Piecewise Refinement of Process
Credit Card Processing
login
Complete Purchase
Buy Ticket
Probably custom
Possible web service
Possible component
19
Successive Schema Refinement
Seat
Event
Program
Composer
Composition
Selection
Available Seats
Musician
1
2
3
Perform
4
5
20
Semantics of Process
Activity (node) describes work to be
done Content state transition, e.g., ?
workspace (e.g., iLog)

Inactive
Active
Pause
Branches control flow from parent to child
execution flow triggered state transitions
information flow sharing of workspaces
21
Product (not business) Opportunities

• Modeling system
• Enable user management of business logic
• Flow execution engine
• High level primitives
• Vertical used in narrow class of apps
• Horizontal used generally
• Solution
• Modifiable and reusable design for specific
  application
• High level vertical primitives usable as
  components and/or services

22
Current Versata Products
Process designer
Incomplete Modeling system


logic designer
Process Engine
Imperfect Flow Engine

Transaction Engine
Transaction Primitives
Horizontal Not very High level

Presentation designer
UI Specific primitives

23
The Software Inferno Revisited
Application
solution
Design
Versata future
vertical
Transaction Primitives
Flow Engine
Modeling System
Versata now
Middleware
Application Servers
Systems
24
Versata Product Strategy

• Project a clear and compelling vision
• based on managing Business Logic (complete
  application lifecycle)
• Strengthen existing products
• as consistent pieces of the vision
• Fulfill the vision
• provide high level primitives

25
Product Development

• Develop an application modeling studio
• Low price, standards driven, wide distribution
• Import as much as possible
• Participate in key application projects
• Develop vertical primitives for domain specific
  applications
• Develop reusable application designs
• Strengthen PLE as flow engine
• Strengthen TLE to support transaction primitives

26
Summary

• A clear vision of where we want to be
• User management of complete application lifecycle
• High level primitives
• Strategy
• Promote vision
• Strengthen existing products within the vision
• Start a long but coherent development cycle