Tuesday and Wednesday at the Lean Software & Systems Conference found me full of interesting thinking and ideas. Almost too full. I have been happy to run into and chat with people that I have known through their blogs or other online interactions. And a few people who I've met in person previously. The tweet stream for the conference went crazy. Oh, and the organization is changing names to Lean Systems Society (announced Monday). Some topics from the day...
Communication. Throughout the whole conference, I heard many people talking about using Kanban and visualization in particular as mechanisms to help open up communication. This is one of the big elements of knowledge work - you can't change it until you can see it. And several people talked about taking the this communication to the next level: change the system, based on the feedback you get from the visualization.
But the visualization is just a representation of the system - not the real thing - as Jim Benson (and others) would remind us. The visualization does a great job of externalizing the pain. But then we have the responsibility of doing something about it: talk about what we see; change the visualization to better represent reality; change the system to reduce the amount of pain. Go. Try. Do. Learn.
Learning. Many of the talks at the conference talked about learning loops and continuous improvement. I was surprised at how many had versions of the improvement loop. I particularly liked that many of them talked about elements where learning occurs. Try something and observe the effects and change based on those observations is a form of single-loop learning. But if you go back and change the model on which you are basing your experiments, then you are doing some double-loop learning. Don Reinertsen warned that there are several different kinds of these loops. Some are focused on speed with OODA as the parent model. Others focus on different elements and get different types of results.
Benjamin Mitchell used the topic of "what comes after visualization" to start a conversation of what to do once you've got some visualization. He particularly talked about Chris Argyris' Ladder of Inference (and expanded by Peter Senge), which he used as a way of thinking about how we see things and how we interact with our colleagues and coaching / consulting clients. He particularly warned about staying away from making assumptions and working at the levels of Select and Describe (rather than Explain, Evaluate, Propose Actions). Since Argyris is one of the promoters of double-loop learning, it is not surprising that Benjamin discussed the Mindset -> Actions -> Results learning loop. I liked the discussion of taking different actions to get results vs changing one's mindset because the Actions aren't getting anywhere like where they need to go.
Jeff Patton had a talk about the myths and other misconceptions that seems to appear in Kanban projects. He was also very proud to have finished his talk on time. He had a top 10, but the highlights seemed to be: Value is hypothetical (in the eye of the beholder); Learning cycles validate hypotheses; Think beyond the edges of the Kanban board. The last item sticks in my craw. The work that happens on Kanban doesn't exist in a vacuum. There are the feeding streams; the coordinating teams; and the processes to whom we handoff our work. They all need help too.
Claudio Perrone described a journey he has gone through with an organization doing Kanban with some interesting stories. One element that was the focus of the talk was that of using A3 Thinking to help solidify descriptions of problems and how the group was going about resolving them. Along with the A3-sized templates, he also seemed to be showing 3x5 cards with quick summaries of the changes to highlight incremental improvements they made to the (Kanban) system over time. I also liked his comment that simply filling out a template (that you get from someone else) is not thinking. Develop what makes sense and move. He closed with a great sentiment - that "Lean is a strategy for reaching objectives through the development of people." and gave us a nice Chinese proverb:
If you want one year of prosperity, grow seeds.
If you want ten years of prosperity, grow trees.
If you want 100 years of prosperity, grow people.
Yochai Benkler (Harvard Law and a bunch of other associations) opened up the day on Wednesday with a great overview of The Penguin and the Leviathan: How Cooperation Triumphs over Self-Interest. His book has been on my reading list for a while, and it has now been bumped upwards - along with several others this week. The discussion of cooperation was based in a lot of research, but the general idea is that research appears to be swinging in a direction that suggests there are elements of self-interest and elements of cooperation built into the way we work. Which behaviors you get depend on a lot of things (nature, nurture, context). But if you work from a basis of "people are selfish" you tend to design a system that rewards individual behavior and punishes deviation. If you work from a basis that people are cooperative and want to help, then you design systems that have different elements (for example: unlimited vacation policies at Netflix and other organizations).
Don Reinertsen spoke about centralized and decentralized control models and how decentralized control could work with examples take from wildfire containment and the US Marine Corps and some anecdotes. Both of these examples focus on fast reaction situations, but I think there were lessons in his talk that extend beyond that context. He described the doctrine that guides these groups, and that sounded a lot like core principles that might apply anywhere. I particularly liked the communication idea that lateral communication is critical to organizations: we need to know what our colleagues are doing and we learn that best by hearing directly from them - not by getting it passed up and down the chain of command. I also liked the reminder that "firefighting" is NOT exemplified by people running around in a chaotic fashion. The underlying situation may be "chaotic" but the responders have to be as level-headed and methodical as possible. I suspect most organizations envisions headless chickens when they think of "fire fighting."
Yuval Yeret asked a version of the question, "Why aren't we doing retrospectives?" This was a good pairing with Karl Scotland and Larry Maccherone's talk on Kanban Metrics. One answer might be that we aren't looking at the right things in retrospectives - taking a scattershot view. Another view might be that we are letting the metrics drive the conversations, rather than the goals (objectives). Both Yeret and Scotland / Maccherone referred to the Golden Circle idea of Simon Sinek - start with the Why and work outwards. Scotland / Maccherone talked about the ODIM model: measurements, insights, decisions, objectives.
And lots of great conversations over coffee, food, couches, and in hallways. I was sorry not to get to do a lightning talk, but I know to craft a better title next time. And maybe there will be some Lean Coffee in Boston to continue these conversations.