#BRF An Evolutionary Perspective of BRMS Presentation

This entry is part 11 of 24 in the series Business Rules Forum 2009

Pedram Abrari of Corticon is presenting.

He started with a quick review of Ron Ross’ definition of business rule and then of some of the reasons why business rule engines came to be.

Shortcomings of first generation BRMS

  • Difficult to use
    • Proprietary programming languages
    • Complex engine algorithms
    • Proprietary rule development infrastructures
    • Specialized debugging techniques to get the rules right
    • The proposed solution: Model driven BRMS
  • Performance bottleneck
    • Deep expertise required for performance tuning
    • Architectural performance ceiling: “Rete wall”
    • The proposed solution: DeTI Algorithm

He then explained the “declarative vs procedural” paradigm and basically said that business rules need to be declarative. Then he discussed the first generation of expert systems evolved from expert systems and that these might not be well suited for general-purpose business rule automation.

Based on a paper from Forrester Research “The truth about business rules algorithm”.

First Generation BRMS:

  • Inferencing algorithms
    • Has authoring flexibility but possible performance issues
  • Sequential algorithm
    • Less authoring flexibility with better performance

Second Generation BRMS

  • Extended sequential algorithms
    • Authoring flexibility with performance
    • Deployment-Time Inferencing, high performance, scalable inferencing

He defined inferencing as the fact that rule engines determine the sequence of rule processing based on rule logical dependencies, allowing more rule authoring flexibility.

Rule engines typically perform two types of iteration

  • Fact iteration (95% of business decision on require fact iteration)
  • Logical loop iteration (rule recursion)

He then performed a short demo of a simplified version of Miss Manners (the benchmark) to explain some of the principles behind “sequential algorithms” and “inferencing algorithms” (and the extended sequential algorithm).

He then covered the evolution of the first generation rule engines from very technical to the additions that have been added since then such as natural language, decision tables, etc.

He also covered some potential problems of first generation rule engines. And obviously he shows how Corticon’s tool solve all these problems.

To me putting down algorithms used by other very successful rule engines to show that your approach might be better might not be the best approach to getting your point across. If the next generation rule engines are about modeling the rules in a new way, then it should be presented as such.

#BRF Vendor Panel: BRMS at a crossroad?

This entry is part 12 of 24 in the series Business Rules Forum 2009

Carole-Ann Matignon of FICO, Scott Irwin of Convergys, Tom Debevoise of Innovations Software Technology and Russell Keziere of Pegasystems.

The session is moderated by Stephen D. Hendrick.

(IBM, JBoss, Corticon, etc. are not there… hmmm…) My notes are not verbatim from the people actually said, but is what I understood as a Gist.

SH: What is the single most important event that needs to happen now in the business rules market?

CAM: We need to understand what happened in the past, we did a lot of progress in the past. We see more analytics, but what we need to get right now is to get the right measurements, better KPIs, better exposed to the business so that you can adapt and improve over time.

RK: Anything that we can do to help visualize and forecast in a reasonable timeframe. Helping PM, Testing, scrum based iteration management. He agrees with CAM.

TD: Truly empowering with the Agility that they need for their rule changes (model, graphical views). Need to make the tools simpler, more intuitive. The technology is an enabler, but not the means.

SI: Rules are moving to the center of an architecture diagram. Being asked to marry Predictive analytics with the rules. More and more event requests as well.

SH: What needs to happen in the next 5 years?

CAM: Same as before. More adaptive systems.

RK: 5 years is a long time. In 5 years we won’t be thinking in application development. We see it with consolidation of the market. Convergence will be happening.

TD: Agrees with RK about the convergence. Need to go beyond the discussions on the specific algorithms. We may be missing the forest from the trees. Move towards Solving the problems from a Process view, event view, rule view.

SI: A number of specifications are evolving.

Audience: Legacy migration and reengineering the systems. What standards are being used or development from portability and interoperatibility so that I don’t build another legacy system

CAM: Many efforts in progress. For BRMS is OMG is working on PRR, but it takes time and very little involvement from customers so that the needs of the customers are more addressed.

RK: Agrees with CAM and talked about something from their products to help.

TD: (Not sure what his point was on that response… Sorry Tom.)

SI: It is key to get customer input because interchange is not high on vendors list of priorities.

Audience: (missed the question)

CAM: Business Rules are built for Agility. Inconsistency between conventional modeling and business rules approach but they can work together.

Audience: Invite vendors to listen to what customers need to resolve problems.

RK: Vendors do listen to their customers and make efforts to try and help resolve those issues.

TD: OMG in PRR is developing a standard modeling for rules similar to BP modeling, but a number of years away. Goal is to make it easy for BA to understand.

SI: Predictive Modeling has some standards, but the specs are slowly moving.

CAM: Reiterates that customers are not involved in standards development so is there a real need? Please help for standards developments

Audience: Social media?

CAM: Community has rolled out about a year ago. I,m a fan of twitter and other things like that.

RK: A couple of months ago we made an announcement about a “sticky note” that you can post and another can “catch”, etc. It is to simplify the process.

TD: Will describe in more detail tomorrow, they are working on “mediated event analysis” in their products.

SI: It is yet another channel for customers to interact. Possibly increasing revenue by mining the information that is out there.

Paul Vincent: What innovations will we see in the next 12-24 months?

CAM: Authoring and exposing the business rules in multiple representation (new kind of representations), managing the KPIs (equivalent to BAM) and how to get all these things working together to empower the user to have a mashup of functionality.

RK: Predicts that … (sorry missed that)

TD: Releasing mediated event service broker. Reverse of Web Service.

SI: Moving to thing integrations with Predictive Analytics, increase scalability and availability.

Audience: Industry Frameworks for rules?

RK: 60% of the business we do has a framework that includes rules. (…)

CAM: We have Best of Breed applications for a widespread set of requirements.

TD:

SI:

To me they did not answer the last question (to my liking anyway).

End of panel

#BRF Difference between CEP and Business Rules

This entry is part 13 of 24 in the series Business Rules Forum 2009

Mike Gualtieri of Forrester is leading the Breakfast with the Analysts session at the Business Rules Forum in Las Vegas.

Confusion between Complex Event Processing and Business Rules

  • Many solutions can be implemented with both approaches

CEP is a software infrastructure that can detect patterns of events from disparate live data sources.

Business Rules

  • Input: Single Payload of Data
  • Conditional logic expressed using multiple metaphors
  • Algorithm: Inferencing, sequential and extended sequential algorithms
  • Output: Data payload or external call

CEP

  • Input: Myriad of data sources all at once
  • Patterns and conditional logic using query or scripting language
  • Algorithms: Stateful pattern detection algorithms
  • Output: Myriad outputs or external call

CEP is unique because of temporal pattern detection… (Note: JRules and Drools actually support this too, but…)

Paul Vincent piped in to say that CEP engines are stateful and can be persisted and in case of failure, recovery is supported as opposed to rules engines which are stateless. (Note: Now this is where the main difference might lie.)

CEP Shines at 3 main things:

  • Detecting patterns and informing
    • Real-time BI
  • Detecting and reacting
    • Real-time trading
  • Event processing architecture
    • Business Process Orchestration

Choose Business Rules if:

  • No need for event correlation (time based)
  • Business Users need to author and maintain rules
    • CEP tools are not as evolved as BR tools
  • Note: I’m adding stateless (i.e. no need for stateful and recovery)

CEP and Business Rules can work together and complement each other.

He mentioned that he thinks that an Event Processing Architecture (EPA) is emerging from the SOA architecture.

Combined CEP and BR are emerging

  • CEP vendors can “poach” the Business Rules market
  • Business Rules Vendors can respond by adding CEP functionality
  • Business Rules and CEP is seeping into other products

His prediction: The future of business rules is CEP. He thinks that BPM is separate and will stay separate.

Good session, I’m not sure I agree with some of the information he presented, for example

  • The fact that CEP can process multiple streams stems from integration technologies which you can leverage around Business Rules Engines
  • He missed the stateful versus stateless brought up by Paul Vincent (which again in most cases stems from integration technologies)

but the ideas he presented were interesting and his prediction of seeing more convergence between those technologies makes sense.

#BRF Keynote BRMS at Cross Roads

This entry is part 14 of 24 in the series Business Rules Forum 2009

Stephen Hendrick is talking on the same topic as yesterday’s vendor panel which he was leading.

Why he thinks that BRMS is at a crossroad (he wont detail all topics):

  • Visibility
  • Passive BRE role in most cases
  • Rule enforcement
  • Limited scope
  • Open source
  • Continued relevance to changing user needs

The market for BRMS in 2008 for product revenue (excluding services) is about 285 million $ and had a growth of 10.5%. BRMS Related technologies is a market of 21845 million $ with 8.2% growth. All software market is 283082 million $ with a growth of 7.0% growth.

He then displayed an interesting chart that presented the vendors on a graph of revenue versus growth. Some forecasts that they are making on the market. He basically doesn’t see any inhibitors to growth and mentioned some growth drivers (market momentum, business awareness, IT spending, Governance, Risk and Compliance).

The forecast is for growth (even in most pessimistic case).

  • Pessimistic: 2% in 2009 to about 7% in 2013
  • Optimistic: 8% in 2009 to about 15% in 2013

He thinks that it will be closer to the optimistic growth in reality based on the signs he has seen recently.

For visibility, before 2008 (purchase of Rule Engines by large ISVs) visibility wasn’t so great, but that is changing. He actually thinks that IBM is in a very good position and so is FICO (although not as good as IBM),. This will drive better visibility for the industry. He also mentions BPMS putting Tibco in the lead because they are well position to execute.

Will open source have an impact? Red Hat and other players may make a difference because Open Source Software is having an impact, the BRMS market is at “risk”. Decision Services and Decision tables can easily be “commoditized”. The authoring tools is where there is a key differentiator.

He then took time to relate Rules and Analytics because rules can be data driven and the analytics (scorecarding, segmentation, data mining) can help in preparing data. Then you can use analytics to refine decisions with transformations, predictive analytics and optimization.

He also linked Rules and CEP because they share some features (active and always on, immediate response, pattern matching, decision centric and data driven). They also complement each other with temporal versus current state approaches.

He used a set of well built slides to present these concepts and their relationships.

He thinks that all of this is going towards a Decision Management Platform (DMP)which combines Events, Process, Rules, CEP, Analytics, etc. He covered some the benefits and challenges for this model.

The cloud is coming. We will see decisions services move to the cloud although some concerns still need to be addressed such as security.

The next 5 years from a vendor standpoint

  • Analytics are coming as a complementary offering
  • DMP is coming by combining CEP, BPM BRMS analytics and EDA
  • Open sources will gain traction
  • Cloud will cater to the more unpredictable performance nuances of DMP
  • (missed last bullet)

The next 5 years from the user standpoint:

  • Analytics
    • Proven to add value to BRMS
    • Hire the right people
  • Decision Management Platforms
    • Monitor major announcements in BRMS, CEP, BPM and Analytics markets

#BRF First hundred days of a BPM effort

This entry is part 15 of 24 in the series Business Rules Forum 2009

Jim Sinur is talking about the benefits of a BPM effort.

BPM efforts have so far a ROI of at least 20%. 84% of BPM efforts are sponsored at CxO level or higher. 60% broken even in less than 1 year. Over 2/3 had benefits over 1 million dollar (record had over 100 million dollar benefit).

The relationship between BPM and rules. To keep your processes green, you need to be able to change your rules in your BPM.

What should the first 100 days of a BPM project look like?

BPM is a journey that affects culture, skill and the way work gets done. Every organization is different and there is no silver bullet.

“Am I going to lose my job?” There is a fear that people are going to lose their job. Need to make people understand that this is going to help them do their job. People feel happier doing their job when structured processes are in place.

Key issues

  • How should users assess organizational readiness?
  • How do you initiate BPM, what does a plan for getting started look like?
  • Critical factors for long term success?

Step 1: Phases of maturity, determine where you are, understand the journey

  1. Acknowledge Operational inefficiencies
    1. Measure and monitor business activities
  2. Process aware,
    1. Model and analyse BP
    2. establish performance metrics
    3. Identify process owners and governance structure
  3. Intraprocess automation and control
    1. Directly link process model and rules to executions
    2. Compare alternatives by various optimization techniques
    3. Integrate activity based accounting with process steps
  4. Interprocess automation and control
    1. Achieve process alignment
    2. Craft process automation across the enterprise customers and trading partners
  5. Enterprise Valuation Control
    1. Achieve comprehensive BRM strategy
    2. Goal driven processes
  6. Agile Business Structure
    1. Innovate new businesses, products and services through agile business structure

Step 2a: Understand your corporate culture. BPM is a cultural change, every organization is different, get help if you don’t have the skillset to do this.

Step 2b: What is your type of enterprise personality profile

  • Cultural aggressive
  • Cultural Moderates
  • Cultural Conservatives

Step 3: Understand Roles and Skill and establish a business process competency center. Critical: Get an executive sponsor.

  • Executive Sponsor
  • Business Process Analyst
  • BP Director
  • BP Project Manager
  • SMEs
  • BP Project Team

Step 4: Executive Sponsor: Get it Right. No absentee sponsors, no iron-fisted approaches, no indecisive leadership. Need to devote the time, be a good communicator, can make decisions, well respected, able to influence and motivated to succeed.

The BPM Plan

  • Determine your BPM Objectives
    • Cost savings?
    • Speed to market?
    • Compliance?
    • Customer satisfaction?
    • Efficiency?
    • Agility?
  • Define your processes and opportunities
  • Determine you initial organizational model
    • Start somewhere, and rapidly evolve to a blended model
  • Apply a “Getting started” checklist
    • (Checklist was too long for me to write all)
  • Document and Use the BPM Strategic Plan

Critical success factors

  • Don’t make the model the primary outcome
  • Plan for a sustainable BPM Program
  • Avoid the temptation to choose the tools first, see the whole, know the right methodology
  • Carefully select the first project
  • Assign the right resources at the right time

Recommendations

  • Understand your culture, political and competitive environment
  • Tailor your approach to meet the unique needs of your organizations
  • Understand the roles and build or hire the skills needed
  • Lay out you BPM strategy and document it in BPM strategic plan
  • Use results to communicate and educate to gain adoption
  • Use BPM to help your enterprise SURVIVE

#BRF Practioners Panel

This entry is part 16 of 24 in the series Business Rules Forum 2009

Miranda Shumaker, Kevin Chase, Kathy Gorman, Mukundan Agaram, Ian Cole, Emily Springer and Mark Myers.

Gladys Lam says that the panellists will introduce themselves, their companies and a couple of do’s and don’ts

Do

  • Hire and train the best BA
  • Adapt the rule methodology
  • Use a rule repository
  • Understand clearly the resource pool available for the tool you are using
  • Talk about decisions with customers, not specifically rules
  • Make sure there is a good mix of business and technical
  • Get executive buy in early on, do evangelism
  • Use external expertise
  • Have a strong support structure in place to support the BAs
  • Keep your rule writing staff together in a CoE
  • Set “Mantras”
  • Define your business language up front
  • Have transparency through the organisation

Don’t

  • Don’t let the business get to involved in the solution
  • Don’t get discouraged, it is a lot of work
  • Start with a pilot but don’t let it get too big
  • Don’t let you techies silo the tool
  • Don’t compromise, due a due diligence
  • Make sure you know which features are available to get your
  • Don’t expect instantaneous success
  • Don’t try to bite of more than you can chew
  • Don’t be afraid to make a mid course correction
  • Don’t be afraid of the word No

GL: What kind of skill sets do BA need to make rule changes, how far you took it, etc.

KC: Look for people that can solve the business problems and that have the logic. “Write those rules as if you own them because you do”.

MA: Good SMEs that can think logically.

MM: (missed the answer)

Audience: BAs belong to business or IT?

MS: BAs should be on the business side but work closely with IT. You should have a Systems Analyst.

KG: BAs are separate from both.

ES: BAs also part of a separate group.

Audience: Stewardship?

ES: Data stewardship with 4 stewards as well as a domain data stewards.

MA: (missed)

IC: Our governance came from compliance, security and these can be drivers to make the call for change.

Audience: Skillset for BAs?

ES: Bus. Architect to model the rule solution. BAs are doing to rule writing. Program office and domains that have BAs trained for rule writing. Skillset: Analytical ability, aptitude to rule writing as mentors for other BAs.

KG: Moving BAs to a Professional BA. Need to change the mindset from being a SME to BAs. Position BAs as expert on the domain.

MM: Soft side of skills. Bridging the gap between IT and the business. Soft skills are key in those people way up there with analysis.

MA: ()

KC: “Get IT out of the process” (Leverage developers on writing code and tools). Mentorships and inspections “Bring everybody up and not tear them down”.

Audience: Business Rules Management, how do you eliminate rules?

MA: Partition your rules into decisions and orchestrating can help.

MM: Rules from the business, you can’t eliminate. But if you had a poor design it might be possible to eliminate rules. Once rules externalized, it is easier to ask “why are we doing that?” and this will help you clean things up.

KC: It’s not changing policy. But refactoring the representation might be an approach to help reduce some of the rules.

Audience: What is the viability of capturing knowledge by business in natural language without involving IT?

ES: “Lobing” over the fence is not an ideal situation. We want BAs to write in natural language and that it will flow to the rule execution side.

MA: presentation yesterday about integrating RuleExpress and ILOG for some level integration. Not perfect but step in the right direction.

MM: The big bang for the buck is on the business side. It is unreasonable to think that all the rules in the system will go in the rule engine. Rules will go in different places. If an error occurs, the system has to show the natural language rule back to the user.

Audience: (missed the exact question)

MS: The technology is not there yet, but hope that there will be this integration between the business pieces and the rule engines. Common language is our small company so IT understands the business language.

Audience: What checkpoint are in place for making sure changes to rules are tracked and OK.

MA: Provided search capabilities and…

KC: If the business makes changes to the rules then it is their responsibility and their accountability. The governance process will help control that. He explained the process they use in their company to help support that (including rule change reports, regressing testing, testing cycle, etc.)

Audience: (missed the question) and most of the answers… sorry.

Audience: Any one using non natural english rules (such as math)?

KC: The rules help support which components of the math calculation you are going to use (which lookup tables for examples) but you are right that some cases are like that.

MS: Some rule are like that, but try to compose them so that it is more closely related to natural language

(missed the last answer)

#BRF Business Rules and Business Events

This entry is part 17 of 24 in the series Business Rules Forum 2009

Paul Vincent is talking about how CEP can help with Business Rules.

His Disclaimer: CEP technologies work alongside other (BPM, SOA, BRE) technologies

Tibco (The Information Bus Company!) is known more for EAI, but has evolved to BPM, CEP, Predictive analytics, etc.

Paul referred to other sessions or keynotes from leading analysts from Gartner, Forrester and IDC from this conference mentioning CEP as one of the key technologies that will be used more in the future.

He did a quick review of what Ron Ross’ definition of rules, or what they can be. CEP becomes interesting for timings and triggers. Chances are it will be a combination of the different types.

Rules are defined through terms and facts, some facts may be events and the rules are enforced as events occur. We want to predict when rules will get broken. It is more important to know that your a heading to bankruptcy, not that you are bankrupt (It’s too late).

Mapping for business rules to processes and decision is easier from an event perspective. He then took some time to how how closely all of the technologies (BPM, Rules, CEP) are closely related and that processes are actually aggregating events.

CEP provides a quicker response to detect issues. It shortens the time between the business event and the action that need to be taken.

Decisions are event driven, CEP identifies patterns in real time. Decisions share data and event so why keep the data in a separate system? Before it was too complex, but now CEP allows to do that to maximize performance. “All decisions are event driven”.

He showed us an example architecture for Real-time event processing. He also presented different high level design patterns that contrast SOA and EDA. A lot of the diagrams presented these concepts and I just can’t blog that kind of information.

Summary

  • Business Rules are used in CEP applications
    • Sense and respond
    • Track and trace
    • Situation awareness
  • Users should model events independently

#BRF Semantic Business Management

This entry is part 18 of 24 in the series Business Rules Forum 2009

Paul Haley of Automata is leading this breakfast session.

Forecasting beyond business rules (5-15 years) these will become indistinguishable in the future. He calls this the Business Stack.

  • Model driven architecture
  • Service oriented architecture
  • Complex event processing
  • Business Process modeling
  • Business Activity Monitoring
  • Predictive analytics
  • Business Intelligence
  • Corporate performance management

A focus on semantics will drastically change the way things will be done in the future. Ontologies will become the model. We will define things in ontologies.

He mentions FOAF (Friend of a Friend) as being a standard for exchanging data.

Business rule realities

  • Derived from AI
  • Primarily based on production rules
  • Substantially forward chaining, goals rarely explicit, no automatic sub-goaling
  • No ability to solve problems or optimize solutions
  • Not enough Ai or operations research

Comment from the audience is that “Constraint Programming does allow for some of the problem resolution that this highlights as not being addressed”, Paul Haley agreed, also “Decision Management generalizes this and includes other technologies which can be used to resolved those problems and that 90% of rules are solved using sequential algorithms as opposed to Rete or others”.

Paul Haley thinks that usually Decision Management is only focused on rules that are part of decision point in a process, but that vision of rules is very limited. (paraphrased)

Paul says that rules in natural language such as “Only full page ads may run on the last page of the Times” but he thinks that computers will eventually ask the questions required to disambiguate the rule statement. He says the “logic systems” will get past this limitation.

Semantic technology (the next step):

  • Semantics – focus on meaning (not structure)
  • Resource Description Format (RDF)
    • Graphs are the universal data structure
    • Metadata is just more data in the graph
    • World-wide identification of nodes, links
  • More powerful, logical deductions
    • Description logic (OWL-DL)
    • Logic programming (Prolog)
    • Predicate calculus (first-order logic)
    • HiLog (higher-order syntax for FOL)
  • More powerful ontology (OWL)

He quickly covered PRR but thinks it has no real added value. OMG’s SBVR on the other hand is very interesting, it is a step ion the right direction but not all the way where he thinks it needs to be. He also mentioned W3C RIF.

He continued by highlighting some of the problems with BI, BOM CEP, BAM, PA, BI and CPM. Basically missing an ontology for all of this.

A Decision is a Process AND an Event. You actually need both.

Ontology needed for:

  • BPMN
  • BMM
  • Rules, Process and Motivation

Sharing ontology across the Business Stack is key.

Somewhat controversial discussion based on the discussions that took place, seems to be an “academic” approach, but still interesting to hear.

#BRF Keynote: Smarter Systems for Uncertain Times

This entry is part 19 of 24 in the series Business Rules Forum 2009

James Taylor his giving the keynote this morning at the Business Rules Forum. No slides, just talk.

Why do you need a smarter system, what is a smarter system?

The pace of change is one of the drivers for creating smarter systems. Changes are so quick nowadays you need to find new ways of doing things.

The competitive environment is also a driver for smarter systems. For example the amazon infrastructure to sell used books is creating competition for the second hand book stores.

Why systems need to be smarter as opposed to smarter organizations and people?

  • Volume: Volume involved in the modern business (transactions or interactions) is so high that there is no way to do this without a system
  • Speed: The amount of data in real time is huge, so only a snapshot is usually looked at because things are happening too quickly. Need to analyze data in real time to identify patterns to respond in real time.
  • Complexity:The environment we work in is getting more complex all the time

What IS a smarter system

  • Action oriented: Not passive supporters. The system needs to enable actions to be taken (either by automating or enabling people to take action) or to make decisions.
  • Flexible (in a business sense): The fact that a business person feels that the system is flexible enough to respond quickly to their needs for change.
  • Forward looking: This links what was essentially Rules driven into Predictive Analytics to peer in the future to help flexible.
  • Learn: A good decision today might be a bad decision tomorrow. Enable you to test and learn an challenge decisions so that you can improve.

How do you get there?

  • It is an approach not a technology platform with a focus on decisions
    • Discover what the processes and decisions are
    • Create decision services. Externalize the rules (Action oriented and flexible)
    • Institute on-going decision analysis to make sure you improve over time

#BRF Business Event Driven Enterprises Rule!

This entry is part 20 of 24 in the series Business Rules Forum 2009

Brian Dickinson of Logical Conclusions is leading this session.

He explains his own definition of events (being an author) and says that he is happy to see how many talks mention events at this conference. Being an event driven enterprise will set you apart from your competition.

An event is anything that happens beyond the are of study that requires or produces a response inside the area of study. – B. Dickinson

All systems have a stimulus-response mechanism. Events make the world go round. In one process view the world consists of a mass of things (events) happening through time. We select the ones we are interested in (sounds like patterns to me).

In the business world and event is what it wishes to respond to from the outside world. The enterprise has no control over theses external customer event, it simply responds to the arrival of the stimuli from the events.

Each external need to which we respond is called an event. An enterprise’s response to an event is a 3 tiered structure: selected events, business analysis, system design and implementation.

There has been a fragmentation of the systems at the implementation which created boundaries at the design level based on the departments, but that is not how things should. The groupings were based on skills of the people that needed the systems.

The boundaries are “historical” (he used hysterical) and using events should allow business to see through these old and outdated boundaries. Engineering the enterprise should be based on events.

Flavors of event:

  • Business Events: to which the business wishes to respond
  • Dependant Events: to which and enterprise responds to satisfy its business events
  • Regulatory Events: To which an enterprise is required to respond

System events are invented at design time. Strategic events are those that dictate the contents of the enterprise’s policy.

Gather 6 things for analysis of events

  • Source
  • Stimulus
  • Processing
  • Memory
  • Response
  • Recipient

You can get rid of “files”  if there is not “intersection” between 2 events.

Let the “data” flow through with no slow down and do your analysis to allow this. The boundaries should not slow down the processing of the events.

If it takes more than “minutes” to get through your company, he says that you are not using event driven partitioning. Don’t combine events, Don’t break events.

Very interesting talk although I wish it would have been longer because we barely skimmed the surface of his topic.