Scrum精益敏捷DevOpsImplications of Technology/技術有話說想法

Implications of Technology:"

2019-05-18  本文已影响30人  StephenTao

Using scientific planning to help you think slowly, many people do not know how to think slowly.

To summarize the P&G approach to scientific planning:1. Frame a choice. Convert each business issue into at least two mutually exclusive options that might resolve it. 2. Generate possibilities. Broaden the list of options to ensure an inclusive range of possibilities. 3. Specify conditions. For each possibility, describe what must be true for it to be sound. 4. Identify barriers. Determine which conditions are least likely to hold true. 5. Design tests. For each key barrier condition, devise a test the team agrees is valid and sufficient to generate commitment. 6. Conduct the tests. Start with the tests for the barrier conditions in which there is the least confidence. 7. Make a choice. Review the key conditions in light of the test results in order to reach a decision. The best part about scientific planning is that it combats cognitive biases through careful experimentation around a broad array of possibilities. Product teams learn to ask challenging questions—questions that designers would find familiar: 1. What are the possibilities? 2. What has to be true in order for a possibility to succeed? 3. What experiments can we run to see if those things are true?
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 95). Pearson Education. Kindle Edition.

The simplest management framework:

Joe discovered that the agile techniques he uses when working with software teams are just the thing for organizing the WIKISPEED crew of a couple of hundred volunteers spread around the world. He found that Scrum—a set of agile practices—gives him a way to establish distributed collaborative teams very quickly and with little overhead, because it provides the minimum set of tools to help team members work well together. He uses Kanban—an agile scheduling technique—to optimize the flow of work within a team. He discovered that principles from Extreme Programming (XP)—especially test-first development—inspire technical practices that create top-quality work. And using principles from lean, WIKISPEED teams are able to maximize the amount of time spent creatively solving problems.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 32). Pearson Education. Kindle Edition.

What conditions are the intuitions of professionals worthy of trust?

Kahneman’s book Thinking, Fast and Slow is the story of our two modes of thinking—System 1 and System 2—which we discussed in the Introduction. To recap, System 1, or fast thinking, is our intuition at work; in most situations, intuition uses a set of shortcuts, norms, averages, or patterns to evaluate the situation and come to conclusions. If we detect that these conclusions aren’t good enough, we switch to System 2, or slow thinking, do our homework, look at the evidence, calculate, and make rational choices.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (pp. 64-65). Pearson Education. Kindle Edition.

The problem is, our slow-thinking self is lazy, so we use fast thinking as often as possible. We don’t really want to know about all of the alternatives that might be available; we would rather just let our fast-thinking mode handle the situation. Actually, this is necessary, because we don’t have the energy to make every decision—or even most decisions—in the slow-thinking mode. But the problem is, according to Kahneman, that we don’t recognize the biases of intuition that influence our fast-thinking self, even when we are aware that these biases exist.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 65). Pearson Education. Kindle Edition.

Cognitive biases make learning difficult, because they make failure hard to detect. We are strongly biased to believe that whatever path we have chosen is the best path and are unlikely to waver in our pursuit of our chosen goal, even if it is clear to others that it is the wrong goal. We are biased to ignore criticism and to be blind to our own failings, even though we are keenly aware of failings in other people. In theory, evidence-based decisions should overcome our cognitive biases. But gathering and examining the evidence requires slow thinking, and slow thinking is hard work, so we try to avoid it. And even when we do examine the evidence, if it contradicts what we expect, our confirmation bias is likely to kick in and explain it away. So even gathering data and doing analysis is not a guarantee that our cognitive biases will be put aside and objective decisions will be made.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 65). Pearson Education. Kindle Edition.

Dealing with Cognitive Biases In the book Decisive, Chip and Dan Heath provide a wide range of techniques to deal with cognitive biases.21 First off, they suggest, important decisions should be approached with an array of options. If you find yourself asking, “Should we or shouldn’t we?” you have narrowed your options to a single choice. If you are making an “either/or” decision, you have limited your options to only two choices. This is the way teenagers are likely to make decisions, according to the Heaths; it is hardly a good way for high-impact corporate decisions to be made. Look for additional options, even options that seem less attractive than the most favored choices. Don’t write off the undesirable options; ask instead, “What would we have to believe in order for these options to be a good choice?” Or ask, “What if none of the options we are considering were available; then what would we do?” In addition to broadening your options, the Heaths recommend that you reality-test your assumptions by running quick experiments and inviting disagreement. They suggest that you step back from the situation and look at it from a distance. Finally, if you are going down a path where most travelers fail—say you are starting a new business—accept the fact that your chances of failure are the same as the chances of those who have gone before you. So it’s a good idea to be prepared to be wrong.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (pp. 65-66). Pearson Education. Kindle Edition.

Klein tells the story of a lieutenant leading a crew fighting a kitchen fire.22 The fire didn’t respond the way it should have—something felt wrong—so he ordered everyone to get out of the building. Just after they left, the floor they had been standing on collapsed into a flaming basement. How did the lieutenant know that something was wrong? He had years of experience observing fire patterns and had built a rich mental model of the various ways in which a house burns. This fire did not match any patterns in his mental model. He realized he did not understand what was going on and became uncomfortable being in the house. So he had everyone get out—just in time. When less experienced people make decisions in equally difficult situations, they use the same approach. But without a rich mental model of the situation, they have limited options to choose from, limited ability to see potential problems while imagining the scenario, and limited ability to detect when the scenario is not going well. The best way to train these novices is to help them to build a rich mental model of the domain by using mentors and simulations. A good way to make sure that novices do not develop into experts, according to Klein, is to expect them to follow standard procedures so that they never make a mistake. For complex, urgent, threatening situations it is much safer to teach people how to recover from mistakes than to focus on making sure that they never make a mistake in the first place. Gary Klein says, “We put too much emphasis on reducing errors and not enough on building expertise.”
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (pp. 66-67). Pearson Education. Kindle Edition.

We need coach:

If you think of achieving mastery in your organization in the same terms as a musician or athlete, you know that for many years, you need a coach. There may come a time when you surpass all available coaches and take on the discipline of deliberate practice yourself, but that is far in the future for most mortals. Coaches, teachers, and mentors play a fundamental role in helping an organization achieve success.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 60). Pearson Education. Kindle Edition.

Indeed, teams need a coach. But we’re saying that the coach will often be the existing manager; in fact, being a coach should be a key part of a manager’s job.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 61). Pearson Education. Kindle Edition.

so we need internal coach and outside coach;

How to define perfection in digital world, do not have failure, or recover quickly and saftly.

Companies that pursue resiliency through failure injection have come to understand two basic facts: 1. Perfection is impossible in a complex system at scale. No matter how low the probability of a failure is, the number of transactions is so large that the only real question is when, not if, a failure will happen. 2. It is a lot less expensive to develop the expertise to recover quickly and safely from failure than it is to pretend that failure will not happen. It can take a radical cultural shift for a company to accept the idea that perfection is not the right goal; resilience is a much better goal in the context of complexity. And resiliency is built on, and builds, expertise.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 69). Pearson Education. Kindle Edition.

How engineers to illustrate the motto "Tribulations are often God's blessings in disguise."

A good example can be found in the discussion “Resilience Engineering: Learning to Embrace Failure.”25 In this article, Jesse Robbins discusses how his firefighter training led him to convince his company (Amazon.com) that the best way to keep their complex systems up and running was not to make them perfect, but to make them resilient. In order to do this, he started GameDay events—events that are triggered by intentionally pulling the plug on a data center (or a similar self-inflicted failure). For a typical GameDay event, it may take dozens of people two or three days to get the system back to normal. Once things are working again, a detailed review occurs in which dependencies and process weaknesses exposed by the exercise are documented and most of them are rapidly fixed. In addition, the people involved in responding to the emergency use the feedback to build their expertise. Resiliency is built over time by breaking things in order to find latent problems and learn how to recover from the inevitable failures that every complex system harbors. This kind of testing reveals everything from very tough problems to simple oversights. Quite a few problems originate in the product development process, so the learning from a single event can move a long way back into the company to change everything from design standards to testing approaches. Other companies with very large data centers—Google, for example—hold similar events. Over time, these exercises have created the learning, confidence, and resilience that allow companies to be comfortable with massive complexity. There is no thought of perfection; it is simply impossible. There is no thought of declaring victory; learning through failure injection is a never-ending journey of constantly getting better. People grow in expertise and gain confidence in their ability to handle the confusion of an outage, and the company’s systems and processes become more hardened against catastrophe.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 68). Pearson Education. Kindle Edition.

High speed VS High quality

most companies believed that speed and high quality were mutually exclusive. However, the CareerBuilder team studied lean concepts and came to the conclusion that this was a false trade-off. They understood that high speed goes hand in hand with serious discipline and excellent quality.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 111). Pearson Education. Kindle Edition.

Inefficient organization in Ericsson

Ericsson had been organized along functional lines, with project managers responsible for delivering product capabilities by coordinating expertise across functions. This structure had a long and successful history, but it resulted in several problems. First of all, handovers of information between functions tended to be inefficient; both knowledge and time were lost in every handover. As the number of handovers increased, the problems tended to escalate nonlinearly. Furthermore, workers in each function were assigned to multiple projects, causing severe multitasking that increased inefficiencies. The inefficiencies of handovers and multitasking showed up as decreased speed, and therefore slower time to market.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (pp. 108-109). Pearson Education. Kindle Edition.

Lean and Agile in Ericsson:

The first step was to change the composition of teams. Existing teams were organized around functions; the new teams would be interdisciplinary teams of about seven people. As they were formed, teams received a brief introduction to the new approach: The program office will track desired features and integration dependencies for products and maintain a continually updated, prioritized list of desired features. Each team will choose (or be given) a feature from somewhere near the top of the priority list. The team is expected to figure out how to deliver that feature. Each feature will have a time constraint (on the order of weeks) determined by its value and market need; the team is expected to finish within that timebox. Features will be described as a high-level goal; teams are expected to decide on the details, sometimes by working with a sales engineer, to decide what scope will meet the goal within the timebox. Team members are expected to decide how they will work together; if they need additional expertise, they are expected to arrange for it themselves. Work must be continuously integrated into the common code base so as to find defects immediately, keep the code clean, and uncover hidden dependencies. A team should work on one feature at a time. Only when it is tested, integrated, and ready to deploy should the team move on to the next feature. Releases will be decoupled from features and scheduled at frequent intervals; any features that are done will be packaged into the next release. The message was clear: “We know at a high level what customers want and when. We expect you to figure out the details of each feature and how to implement it, and we will support you.”
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (pp. 103-104). Pearson Education. Kindle Edition.

What Is Efficiency?

The most efficient retail stores—the stores that can afford to offer customers the lowest prices and best values—are the stores that pay the highest wages, offer the best benefits, and provide stable, full-time employment.1 The most efficient U.S. airline—the one that has the lowest labor costs per passenger mile—is the one that pays employees the highest salaries and has never had an involuntary layoff.2 The most efficient customer service centers—the ones that have the fewest calls and the most satisfied customers—do not focus on worker productivity; they focus on understanding the causes behind customer calls and on eliminating those causes.3 The most efficient product development organizations—the ones that design the most valuable products—are not the ones that deliver the most features, but the ones that deliver only the right features.4 Product development efficiency is not about skimping on product development costs; it is about focusing creative minds on the job of developing successful products. The only viable measure of development efficiency is the ratio of effort expended to overall product performance. In their excellent book This Is Lean: Resolving the Efficiency Paradox,5 Niklas Modig and Par Ahlstrom distinguish between resource efficiency (keeping everyone busy) and flow efficiency (keeping the work flowing). They show how an intense focus on resource efficiency introduces intractable wastes such as queues, handovers, and thrashing. Truly efficient organizations focus primarily on flow efficiency, which provides a holistic view of the system and inevitably produces better overall results.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (pp. 101-102). Pearson Education. Kindle Edition.

the evidence clearly shows that practices that demoralize workers make it impossible to have genuine, sustainable efficiency. Lean leaders understand that true efficiency comes from flow, from speed, from learning, and from workers who care.
---Poppendieck, Mary. The Lean Mindset: Ask the Right Questions (Addison-Wesley Signature Series (Beck)) (p. 102). Pearson Education. Kindle Edition.

上一篇下一篇

猜你喜欢

热点阅读