Competence Levels and Intrinsic Motivation

I wanted to share a bit on ideas coming from NLP (for Neuro Linguistic Progamming). These are ideas that my wife practices (so, I am somewhat irradiated by the way). There is also a relationship with creativity workshops I followed in the past. My wife is working on how to apply strategies that work, how to model behavior. Her scope is on how to teach kids about strategies that work. She's pretty good at that. So, when we model successful people, we find that they have strategies that work and if we apply them, it will work for us too (usually it boils down to working hard and smart and having a good network, but it helps having some clues on the details 😉 ). In that respect, when we learn something, there are *always* those 4 stages present:
  1. Inconscious Incompetence
  2. Conscious Incompetence
  3. Conscious Competence
  4. Inconscious Competence
 This is also linked to what I call the valley of dispair: When someone moves to the conscious incompetence level, a whole new field of knowledge opens and it is usually pretty frightening. The time it will take to master the field is unknown and this generates great stress. This challenges the homeostatis of the being. So, there is a tendency to step back and stay off the new field (easier to avoid pain than to work towards gaining pleasure). Add to this the external environment where people who are still at the unconscious incompetence level always try to drag that someone back to them... Furthemore, learning the new skill takes energy and is usually less effective than doing things the way they always have been done in the organization (at that moment, the individual is experiencing the dispair, doubt about the new way is experienced etc). No wonder that it's difficult to decide to learn for good despite those calls. From there, several strategies are available : 1) keep on pushing like mad ("I'll do it, I'll do it, ... whatever it takes) This usually leads the individual to burn out and shoot the messenger kind of reaction towards the person. The person cannot stand staying on a plateau level for too long and call quits. 2) jump to another topic constantly This way, the conscious competence level may sometimes be reached by luck but usually leading to a dillettante style. Knows a bit about everything, brags, talks all the time etc. No quantum leap in personal performance. 3) decide to be intrinsically motivated and walk the talk on real problems, enhancing the practice believing that the time will come without becoming a bigot of the practice. Conscious competence builds up and at one moment, the person realizes that a jump to inconscious competence happened. (Without noticing, usually opening new perspectives). As an additional tought, when knowledge increases, it's like the radius of a circle. And the unknowns increases like 2 * PI * R ... The more one knows, the more unknowns there are. But, the person becomes much more sensitive on a lot of subjects. As a sample case, look at the career of Peter Coad where he started on design [handbook], moved to more conceptual things [archetypes], worked with a vision [together], and then know works as a strategist (environment=design, beliefs=vision, transcending|identity=strategy). [He's one of my role models as is Jim Rumbaugh whose OMT is the first method I used to move away from messy arrows and boxes, and Grady Booch with the "Managing the OO project" book. Let's not forget Ivar Jacobson obviously, and Alan Weiss, a great mentor.] I guess that this is a natural path to what some call "enlightment". Maybe can we reach some kind of it through software engineering or "architecture" of systems (like 4+1, form & function as you use in some talks). One of the rules of NLP is that the mind and body do form a cybernetic system (which is also function and form). In my courses, I find myself more and more talking about beliefs and what would be the business case for UML (benefits vs costs). This makes very interesting talks. This may be a consequence of the unconscious competence mode... starting bending the rules and asking questions, pushing the boundaries, understanding the trade offs and beliefs that were in the minds of the creators. Any similar experience of yours?

Sparx EA – Modeling Business Processes with BPMN and EA : A case study of Callataÿ and Wouters

Albert and BPMNBPMN basics are the next topic in the session. The point of model a business process is to capture the organizational way of working. Typically we talk about operations, customer support, and R&D. A notation is needed to talk with each other. BPMN provides a visual notation. It is standard now. Enterprise Architect supports BPMN1.1. The OMG (Object Management Group, altough sometimes they err on the side of « Oh My God » on the complexity front) is maintaining it. Well, actitity diagrams do form the basis for modeling the processes, made of activities, decisions, transitions and so on. So, why BPMN when we have activity diagrams ? BPMN provides a much richer notation, with cleared building blocks (receiving, complex decisions, resources). FYI, there is a core set of diagram elements, namely : Event, Sequence Flow, Activity, Message Flow, Gateway, Association. In the complete set of diagram elements, we ge tthis like special event types : message, timer, exception, cancel, compensations, rule, link, terminate, multiple. Activities also get a fair treatment on that front. Subprocess, task, recurrent task... We can also use Pool et Lane. Pool is a pool of people (think and organization). Inside a pool, you can get multiple lanes. One person can be a lane as well. Some rules do apply. Message flows can only go between pools, because you are not going to control the business processes of other organizations (they won't let you!). All of these things are depicted with a nice little icon on the diagram. This would be a bit short but there are also the tagged values and the specific pick lists in the tagged values windows for supporting that. All in all, the BPMN support is very decent. As usual, EA provides top notch document generation facilities and getting printouts that hold water nicely is a non-issue and work very well. I jumped in (could not resist) and explained a bit about how BPMN and EA work well hand in hand along with the quicklinker, the shortcuts for creating the same item over and over (shift-F3, F3) if you have lots of them, the fact that you can put the tagged value window docked in a nice place. The tagged value window is crafted in such a way that it behaves in an improved way when dealing with BPMN. Case study time ! Mr Pierre-Philippe Delmarcelle presents the case study. He is Enterprise Architect for C&W (I would say tought leader on these issues). Mr Delmarcelle from Callataÿ and Wouters (2) Callataÿ and Wouters faces the challenge of having customers asking about how to solve business issues and not only be a technical provider. Internationalization brought its own challenges : standardization needs(BPMN helps when working with integrators, consulting, ...), explaining how the solution works. Traceability is very important due to regulations. With risk reduction needs, customization is not that hot anymore. But flexibility points are needed still. To support Enterprise Architecture, we need a business process model, an application architecture model and a systems model. The business is modeled with the Business Object Model, the Business Landscape (BIAN aligned), the Business Process Model, the Business Organization Model. These things have been (quickly) created in the previous months of 2009 with the help of Expert-IT. TEA is the name of the enterprise architecture of the Thaler product of C&W. Traceability between the process activity and the components that realize it is maintained. A number of business areas are established (the diagram appears quite busy, showing the complexity of the undertaking). So, this screen supports this or that process, being part of a specific domain. All modeling was done using BPMN in Enterprise Architect. The full metamodel has been created in Enterprise Architect and made available as a custom toolbox, a set of rules governing the model and so on. Analysts do work fully inside Enterprise Architect and output RTF documents from which PDF document can be delivered. Import/export of model parts is done with the XMI (Package Control features) so that analysts can work in a disconnected way. Archimate serves as a solid metamodel as an inspiration source. EA is a very nice tool for Pierre-Philippe Delmarcelle:very flexible, not costly, with lots of exports (XMI allows to move it into other tools (mega, aris)). How to use the business process model in the projects. Before : Customer Questions : How will you do the implementation? Are there gaps? Since time to market is money, this is helping tremendously on the added value front. Before, we conducted demos and looked at screens. This was very fine grained, missed the big picture. Now (with the new methodology) : the thinking with landscape model helps in scoping faster the discussion. With the client, we can now come with our Thaler reference business model in the initial workshops. The business processes of Thaler and the ones of the customer can be compared. We can now explain how our solution works in business terms. Please note that the business processes in the blueprint are clearly highlighting how Thaler shines as a solution there. Some demos are still necessary but much less than before. The outcome of the workshops provide clear guidance on how to perform the implementation in a value-added way, with a high probability of success. The gaps are now made clear and can be addresses with an economy of means. An outcome of the new ways is that the responsibilities are much clearer when it comes to the customer side and the integrator side. The improved quality is a key addition. As a side effect, this brought a new service when it comes to helping in tailoring business processes. Clear communication is really the power of business processes. Lot more gaps identified with the new approach (up to more than 278 in a large project. Reduced to almost zero when looking at the ways that Thaler standard system worked!). Risks are really going down. Key take aways :
  • TEA is now the underpinning of everything
  • A guiding blueprint is useful for :
    • Client facing demand management to create a solution enabling strategy
    • Client facing implementation much less risky and aligned with the strategy
  • Customer and internal projects are accelerated, risk is mitigated
    • Better communication
    • Start form business concepts
    • Use market agreed concepts leading to consensus fast
  • Efficiency and cost reduction achieved
    • Better scoping of requirement and management of development
    • Standardization and resuse of common concepts
      • Enable SOA strategy for the solution
      • Without the bluepring and details, SOA cannot be truly realized because you need a catalog of services that do make sense in real situations
This makes it very clear that the Enterprise Architecture is not only vaporware but a pragmatic, value-deliverying, business accelerating approach. Callataÿ and Wouters proves it and as such is positioned as a market leader in its field. Kudos to Pierre-Philippe Delmarcelle and team for his great presentation and great vision. His achievements of creating more than 120 processes, defining the methodology, producing the reporting, training teams and partners has been defined in less than 6 months. BIAN proved to be a great asset when it comes to mastering complexity and chunking it in manageables pieces that can be assimilated properly. As a bonus, TEAL -- Thaler Enterprise Architecture Language - a special BPMN evolution has been defined, implemented and put to use. Sparx Enterprise Architect served as the central repository that federated everything together. So: BIAN + TEAL + Blueprinting + Components + Enterprise Architecture = Success and Solid Basics! Like in the A-Team, "I love it when a plan comes together".

Sparx Systems EA – new 7.5 features

Phil & Jean-GabI attended a session on the new Sparx Systems Enterprise Architect organized by my friends at Expert-IT. This time, I am in the audience and not presenting anything. Last time, I was showing Ivar Jacobson Essential Unified Process and EssWork. With the new 7.5 version, Sparx systems Enterprise Architect, there is a move further than structural code generation. From now on, STD, sequence, and activity diagrams can be used to generate code. In fact EA now moves into new ground: mindmapping, business processes modeling, org charts. For business users, BPMN is a good way to capture the processes in a strong repository. A forte of the tool is the traceability features. The fact that EA is based upon a strong relational repository helps tremendously on achieving that. Impact analysis is made easier. Also, business rules are now a first class citizen, more on that later. All artifacts|work products can be stored in the repository as well. EA 7..5 is a major release. No 8.0 yet but we can expect a lot more exciting characteristics end of year. New pricing, albeit still reasonable. What is new is that additional editions are available. Business and Software, Systems Engineering, Ultimate Edition. Since the customers attack larger problems with the tool, performance is improved for WAN communication. WAN optimizer is available as well as a lazy loading feature allowing to load data on an as needed basis. You'll need Pro for the lazy loading and Corp for WAN optimizer. On the Business Process modeling front, the beast can generate BPEL from BPMN 1.1 (get your acronym dictionary ready!). DoDAF, TOGAF, MODAF and Zachman Framework are included. There is also Archimate somewhere in the mix When UML is lacking, EA provides the missing pieces, a good point. MDG integration and MDG link are not the same things. About the Systems engineering edition now: SysML and simulation are in, and code generation to hardware languaged works: Verilog and friends. BPEL support takes the form of additional dialogs for example. BPEL is drawn in BPEL diagrams using the BPMN icons. Please stay aware that BPMN is for the 'what' of a process and BPEL is 'what' since we want to get XML for running inside a process engine. The new business rules system is on steroids: a fact model (a kind of class diagram allowing to capture facts that work with the new feature) is created and an 'orchestration' class is needed there. That special class then gets support from an associated ruleflow diagram for process-type rules (this time an activity diagram retooled a bit for BR support). Ruletasks are the activities there. Moving further, a business rule dealing with decisions is present and uses the elements of the fact model to author clean rules in a rule composer. Nice! Then code can be generated by the tool. This is clean and allows bridging the developers and the business analyst. When it comes to impact analysis, this is great. The challenge is to work at the right granularity level where the rules under modeling are going to offer a clear benefit/investment ratio. As always, model with a purpose. Maybe you want to do some simulation. Build and run is inside EA now, so, self containment is achieved. Another acronym: CTF is of use (Code Template Framework). Strategy Maps, Balanced scorecards, value chains. Flow charts, decision trees and orgcharts do complete the new features. This can move EA inside the boardroom in my view. Obviously as an output, not asking C-level people to draw them. But who knows... Scripting is now fully doable inside an EA code editor. And the files are part of the project if you want to do so. Great! Debugging is also available. VBscript and JavaScript are supported. Intellisense works and helps while navigating the object model of Enterprise Architect. An advanced math library is available. No clue on what it provides yet. More advanced wizardry with EA is available with MDG technologies. New toolboxes, items, ... The sky is the limit (well, as always do not confuse the means with the ends. Falling in love with tooling and ending up with a new addiction is not going to put bread on the table).