Laws of Software Engineering

Frameworks

Ways of seeing the whole picture. They give you better questions to ask.

How to use them. Use one when you are stuck because the problem is the wrong shape. Cynefin asks "what kind of situation is this?" Wardley asks "where are we, and where is this moving?" Theory of Constraints asks "what is the slowest part?" The framework changes the question, and a better question changes the work.

How they get misused. Treated as a summary. Nobody learns a framework from a one-line description. The entry points you at the framework. The work is reading the source, practising the move, and getting it wrong a few times.

• Framework Wardley Mapping Plot the things your system depends on. Two axes: how visible they are to the user, and how mature they are. Watch how they move over time.
• Framework Deming's System of Profound Knowledge Four ways of looking at a problem that work together: seeing it as a system, understanding variation, asking how we know what we know, and paying attention to people.
• Framework Cynefin Five kinds of situation: clear, complicated, complex, chaotic, and confused. Each one needs a different response.
• Framework Theory of Constraints The speed of any system is set by its slowest part. Improving anything else makes no difference.
• I Framework Improvement Kata A four-step pattern for getting better at anything. Understand the direction. Grasp the current condition. Set the next target condition. Run experiments toward it.
• J Framework Jobs To Be Done People do not buy products. They hire products to do a job. Understand the job.
• T Framework The Three Ways A way of seeing DevOps as a whole. Optimise flow from idea to production. Amplify feedback from production back to development. Build a culture of continuous experimentation and learning.
• T Framework The Fifth Discipline Systems thinking, personal mastery, mental models, shared vision, and team learning. The five disciplines of an organisation that can learn.
• L Framework Lean Enterprise An operating model for large organisations that want to be fast. Optimise the whole value stream. Build feedback into every step. Treat the business as a portfolio of experiments.
• B Framework Beer's Viable System Model A model for designing organisations that can survive and adapt. Five interdependent systems handle operations, coordination, control, intelligence, and identity.
• S Framework System Dynamics A way of modelling systems made of stocks, flows, and feedback loops. The behaviour of a system over time follows from its structure, not from individual events.
• W Framework Wicked Problems Problems where the problem statement, the solution, and the evaluation are all contested and interdependent. There is no stopping rule. Every attempt at solution changes the problem.