18. The New Definition Of Success

Martin Thompson on Arrested DevOps, Dr. Carola Lilienthal on Legacy Code Rocks, Jeff Gothelf on Agile Atelier, Safi Bahcall on Coaching For Leaders, and Mike Burrows on A Geek Leader.  I’d love for you to email me with any comments about the show or any suggestions for podcasts I might want to feature. Email [email protected]. This episode covers the five podcast episodes I found most interesting and wanted to share links to during the two week period starting August 19, 2019. These podcast episodes may have been released much earlier, but this was the fortnight when I started sharing links to them to my social network followers. MARTIN THOMPSON ON ARRESTED DEVOPS The Arrested DevOps podcast featured Martin Thompson with host Jessica Kerr. Martin and Jessica talked about the parallels between optimizing the performance of software systems and doing the same for human systems. Using ideas from queuing theory, they discussed the notion of adding small amounts of slack to a system to make it drastically more responsive. Martin connected Amdahl’s Law to the more general Universal Scalability Law, which is more comprehensive because it takes into account coherence cost, which is the time needed to reach agreement between parties working together. He added that Brook’s Law from The Mythical Man Month is the Universal Scalability Law by a different name. They talked about the difference between parallelism and concurrency. Parallelism, Martin says, is doing multiple things at the same time. Concurrency means dealing with multiple things at the same time, a definition Martin says he stole from Rob Pike. He further decomposed the universal scalability law into its parameters. One parameter represents whether you can subdivide the work (the contention penalty) and the other represents the time to reach agreement (the coherence penalty). If your team can reach agreement faster, they can get better throughput because they can have more parallelism with less concurrency. They got into a discussion of the importance of feedback in information theory. Sending information and not confirming reception is a naïve approach and this has been understood for a long time and yet software is still built that ignores this. Two phase commit is an example. If you study the two phase commit protocol in any detail, Martin says, you realize it is fundamentally broken, yet corporations don’t want to say that. They talked about how to design distributed applications in the presence of partial failures. Martin says to make your communications idempotent, give each message a sequence number, and use this sequence number to identify and ignore replayed messages. According to Martin, designing your systems this way is just good hygiene and professionalism. Apple Podcasts link: https://podcasts.apple.com/ca/podcast/protocols-and-sympathy-with-martin-thompson/id773888088?i=1000444947737 Website link: https://www.arresteddevops.com/protocols/ DR. CAROLA LILIENTHAL ON LEGACY CODE ROCKS The Legacy Code Rocks podcast featuring Dr. Carola Lilienthal with hosts Andrea Goulet and Scott Ford. They talked about Domain-Driven Design. Carola said her company read Eric Evans’ book and immediately took to it. Talking to users, writing software in the user's domain, and using a common vocabulary fit with what they were already doing so they adopted it easily. They talked about Carola’s modularity maturity index. It consists of three areas of sustainability: 1) modularity, 2) hierarchy, and 3) pattern consistency.  Andrea brought up the fact that larger codebases aren’t necessarily more difficult to change as Carola found in her research. Carola says that, based on the three hundred systems she’s studied, systems under a million lines of code are often in a worse state than larger systems. Around a million lines of code, she says, something happens: either people start structuring the system and putting in guard rails that keep the product maintainable or the system doesn’t grow any more. Apple Podcasts link: https://podcasts.apple.com/ca/podcast/sustainable-software-architecture-dr-carola-lilienthal/id1146634772?i=1000443349633 Website link: http://legacycoderocks.libsyn.com/sustainable-software-architecture-with-dr-carola-lilienthal JEFF GOTHELF ON AGILE ATELIER The Agile Atelier podcast featured Jeff Gothelf with host Rahul Bhattacharya. Rahul and Jeff talked about the intersection of Agile, Lean, and Design Thinking to find commonalities. They examined customer-centricity, measuring success, continuous testing, and the importance of having a hypothesis. Jeff had been working as a designer on waterfall projects for the first decade of his career and, on a good day, only saw 50% of his work get implemented. Ten years into his career, Jeff got exposed to Agile software development and it forced him to revisit his design process and his process for doing product development as a whole. Because Jeff was in a leadership position and had a boss that understood the new methodology, Jeff got the chance to run process experiments to learn what the best collaboration model was for him and his team. This became the basis of his book, Lean UX. Rahul asked Jeff how he would define Design Thinking. Jeff described Design Thinking as applying the designer’s toolkit to solve business problems. This includes empathizing with customers, brainstorming ideas, prototyping, testing ideas with customers, and iterating.  Rahul asked if there is a specific situation in which to apply Design Thinking. Jeff says that he has yet to find a client or an industry where customer-centricity, continuous learning, risk mitigation, experimentation, and iteration don’t make sense. Even when working with people at GE who make locomotives and working with organizations that make room-sized air conditioning units that sit on top of skyscrapers, Jeff was able to successfully introduce them to ideas like talking to customers, identifying risks, and continuously improving their product. Rahul asked how the principles of Design Thinking fit with the Agile principles. Jeff says that everybody thinks that Agile is its own thing, Design Thinking is its own thing, Lean Manufacturing and Lean Startup are their own thing. The tactical execution of those methodologies might be different, but at their core, Jeff says these methods all share the same principles.  They are all customer-centric. They all measure success as an outcome, as a change in customer behavior. They all focus on testing your ideas quickly and moving off of bad ideas quickly. And they all focus on continuously improving and iterating the thing you are making as you continue to invest in it. They then got into a discussion about the importance of measuring the impact on the user of the product you are building. Jeff says that, unfortunately, shipping the thing is still one of the major definitions of success for most organizations. But in a world of continuous software when you can push a software update five times a minute like Amazon does, delivering the thing is a non-event and it should be a non-event. We shouldn’t celebrate it. What we should celebrate is the change in customer behavior that tells us that we’ve delivered value. These are things like showing up at the website, engaging with the app, buying the product, telling your friends, whatever it is we care about for our product. This line of thought led to the quote above. Apple Podcasts link: https://podcasts.apple.com/ca/podcast/episode-11-intersection-agile-lean-design-thinking/id1459098259?i=1000445718430 Website link: https://rahul-bhattacharya.com/2019/07/30/episode-11-the-intersection-of-agile-lean-and-design-thinking-with-jeff-gothelf/ SAFI BAHCALL ON COACHING FOR LEADERS The Coaching For Leaders podcast featured Safi Bahcall (author of the book Loonshots) with host Dave Stachowiak. They talked about what science has to say about the best ways to nurture new ideas. They started out with a discussion of children’s books and Safi’s first example of a loonshot was Dr. Seuss. He had just been rejected by every publisher he took his first story to when he ran into a friend in the street. This friend asked Dr. Seuss about what he had under his arm and when he found out it was a manuscript for a children’s story that Dr. Seuss was taking home to burn, the friend revealed that he had just taken a job at a publisher across the street and asked Dr. Seuss if he would like to come into the publisher’s office. The Cat In The Hat was born. Safi used the story of the moon landing as an illustration of the difference between a moonshot and a loonshot. A moonshot was Kennedy’s speech announcing that the United States would put a man on the moon by the end of the decade. A loonshot was forty years earlier when Robert Goddard suggested getting to the moon with liquid-fueled jet propulsion and was ridiculed by many, including the New York Times. The reason it is important to understand the difference is because Goddard’s ideas, though neglected by the Americans, were embraced by Nazi Germany. German scientists used Goddard’s ideas to build jet engines and planes that flew 100 mph faster than any Allied plane. The mistake of neglecting Goddard’s ideas was fatal. Companies often ask Safi how they can innovate and create new products while continuing to keep their original product or service competitive. He thinks about these situations using three metaphors: the ice cube, garden hoe, and heart. He starts by thinking about the artists who create new product ideas and soldiers to execute on turning those ideas into real products in the marketplace. The ice cube is a rigid phase that suits the soldiers and a melted ice cube is a fluid phase that suits the artists. Understanding the problem starts with the ‘beautiful baby’ problem. The artist sees their new idea as a beautiful baby. The soldiers look at the same thing and see a shriveled up raisin. They’re both right. The garden hoe comes from understanding that the failure point in most innovation is rarely in the supply of new ideas, it is in the transfer between artists and soldiers. Great leaders are those who think of themselves as gardeners managing the transfer between the artists and soldiers. The heart is about loving your artists and soldiers equally. When we lionize the artists as the media often do, we demotivate the soldiers. I liked what Safi had to say about the problem with following the standard advice about active listening. Apple Podcasts link: https://podcasts.apple.com/ca/podcast/418-the-way-to-nurture-new-ideas-with-safi-bahcall/id458827716?i=1000443895174 Website link: https://coachingforleaders.com/podcast/nurture-new-ideas-safi-bahcall/ MIKE BURROWS ON A GEEK LEADER The A Geek Leader podcast featured Mike Burrows with host John Rouda. Mike talked about his career leading up to the writing of AgendaShift. He described the goal of AgendaShift as trying to introduce agility not by prescribing a set of practices or rolling out a framework but by getting agreement on outcomes and working out different ways of achieving them in an hypothesis-driven way. He then mentioned his newer book that he was working on at the time the podcast aired and has just come out this month, Right to Left. Right to Left is about working backwards from outcomes. John asked what the shift was that led to this outcome-focused approach. Mike said that while working in the government digital space in the UK, he witnessed rapid change. Instead of one supplier creating documentation for a new system, a second supplier building it, and a third supplier supporting it, and the whole thing being an expensive mess that disappoints its end users, he says they now have a system where projects will be halted if they are not serious about engaging with users, doing user research, understanding needs, and working iteratively to deliver evolving services. He says that if it can happen in the government space, it can happen anywhere. John asked about what a new manager coming from an individual contributor role would need to learn for dealing with the people side of managing projects. Mike recommended tempering any temptation to micro-manage. On his first day taking over a management position at UBS, he had people lining up at his desk looking to be micro-managed because that is how his predecessor worked. He told them that if this is how it is going to work, it is going to make him miserable and it is going to make them miserable and he encouraged them to self-organize. Mike’s second recommendation is to learn to value and respect people who come from other disciplines than technology, as he says in the above quote. John asked Mike to describe AgendaShift. Mike says that the best two words that describe it come from Daniel Mezick: it is an engagement model. Much like Daniel’s OpenSpace Agility, AgendaShift describes how change agents can engage with their organizations. In the Lean/Agile space, pushing Agile on people is self-defeating and creates more problems than it solves. Instead, facilitate outcomes that the people of the organization can agree on and start solving problems. AgendaShift starts with discovery. There are workshop tools to creating a high-level plan. Then they use an assessment tool for identifying opportunities to increase transparency, get workloads under control, or to engage better with customers. They identify obstacles and the outcomes hiding behind those obstacles. They use a “clean language”-based game to model a landscape of obstacles and outcomes and get people to think about the journey, their priorities, and what the key landmarks along the way will be. Apple Podcasts link: https://podcasts.apple.com/ca/podcast/agl-081-agendashift-with-mike-burrows/id1043194456?i=1000424584602 Website link:https://www.ageekleader.com/agl-081-agendashift-with-mike-burrows/ LINKS Ask questions, make comments, and let your voice be heard by emailing [email protected]. Twitter: https://twitter.com/thekguy LinkedIn: https://www.linkedin.com/in/keithmmcdonald/ Facebook: https://www.facebook.com/thekguypage Instagram: https://www.instagram.com/the_k_guy/ YouTube: https://www.youtube.com/c/TheKGuy Website: https://www.thekguy.com/ Intro/outro music: "waste time" by Vincent Augustus