Parveen Kaler

Book Notes: Ultralearning by Scott Young

Mon, 24 Aug 2020 00:00:00 +0000

Ultralearning is a prescriptive work that teaches an intense and self-directed strategy for acquiring skills and knowledge.

Principles of Ultralearning

There are 9 principles in Ultraleaning:

Metalearning
Focus
Directness
Drill
Retrieval
Feedback
Retention
Intuition
Experimentation

Metalearning involves researching syntopically. Start by drawing a map of how to learn the subject or skill. Metalearning may differ based upon Instrumental vs Intrinsic projects. Instrumental projects are those with the purpose of achieving a specific result. Instrinsic projects are those that are pursued for their own sake.

Interviewing an expert may help with Metalearning. The key is to write a simple, to the point email, explaining why you’re reaching out and asking for fifteen minutes of their time.

The subject can then be structured into Concepts, Facts, and Procedures.

Focus involves cultivating time to concentrate.

Directness involves learning the subject by directly working on the subject rather than proxies. Opting for a project rather than a class may be better for learning. Immersion is the process of surrounding yourself with the subject or skill. This allows for much more practice than typical.

Drill involves improving your weakest points by repeatedly practicing. Careful analysis and deliberate practice is key. This is especially key for the Rate-deterimining steps.

Retrieval involves actively recalling information. Flash cards are a simple but effective tool.

Feedback involves calibrating knowledge of the subject or skill. There are three types of feedback: Outcome, Corrective, and Metafeedback. Learning rate is an example of metafeedback. High-intensity, rapid feedback provides both informational and emotional advantages.

Rentention involves learning things for now and forever. Spreading learning sessions over more intervals over longer periods of time tends to cause better performance in the long run.

Intuition involves getting a feel for the subject by exploring and playing with concepts. Most people learn abstract rules after being exposed to many concrete examples.

Experimentation involves discovering past the known edges of a subject and discovering new knowledge.

Book Notes: The Score Takes Care Of Itself by Bill Walsh

Thu, 25 Apr 2019 00:00:00 +0000

The Score Takes Care of Itself describes the attitudes and actions required for implementing a high Standard of Performance.

Standard of Performance

Guidelines for createing a Standard of Performance:

Create a comprehensive list of specific attitudes and actions
Communicate clearly that high effort and execution is expected
Communicate expectation of high expertise
Teach connection and extension to the team
Make metrics explicit

Fundamental leadership assertion: Regardless of your specific job, it is vital to our team that you do that job at the highest possible level in all its various aspects.

A philosophy is the aggregate of your attitudes toward fundamental matters and is derived from a process of consciously thinking about critical issues and developing rational reasons for holding one particular belief or position rather than another.

Hire personnel with four characteristics:

Talent
Character
Functional intelligence
Eagerness to adopt philosophy

There was no timetable for the 49ers to win a Super Bowl. There was a comprehensive standard and plan for installing a level of higher competency in all areas. Have faith the score would take care of itself.

Innovation, Planning, and Common Sense

Create scripts or plans before heading into “foul weather”:

Flying by the seat of your pants precedes crashing by the seat of your pants.
When you prepare for everything, you’re ready for anything.
Create a crisis-management team that is smart enough to anticipate and plan for crises.
All personnel must recognize that your organization is adaptive and dynamic in facing unstable weather.

People, Priorities, and Performance

Checklist of personal qualities in staff members:

A fundamental knowledge of their area
High energy and enthusiasm
Ability to discern talent in potential employees
Ability to communicate
Unconditional loyalty to team

Checklist for keeping staff on the same page:

Establish clear parameters and methods for staff
Address philosophical differences privately and one-on-one
Use different approaches that vary with staff
Be visible to all departments
Delegate but monitor
Correct differences in philosophy in different departments
Watch out for teaching of personal beliefs
Prioritize personal contact

Daily reminders to stay on track during a losing streak:

Concentrate on what will produce results rather than on the results, the process rather than the prize.
Rely on expertise, composure, patience, and common sense
Maintain the Standard of Performance
Don’t isolate yourself
Don’t let the magnitude of the challenge take away from the next incremental step.
Stay upbeat and determined.
Hold staff meetings with what to expect.
Don’t label a new plan as the thing that will “get us back on track”.
Maintain courtesy and respect.
Don’t plead with employees to “do better”.
Avoid threatening or chastising.
Deal with immediate superiors one-on-one.

Book Notes: High Output Management by Andy Grove

Sun, 21 Apr 2019 00:00:00 +0000

I am writing a series of posts on management books including High Output Managment, Measure What Matters, The Score Takes Care Of Itself, and Trillion Dollar Coach.

Subscribe by RSS or Email to get updates.

High Output Management is concerned with improving the performance of technology company managers. It contains three central ideas:

An output oriented approach to management.
The output of a manager is the output of organizational units under his or her supervision and influence.
High productivity depends on performing high leverage tasks.

Leverage and the Factory Metaphor

The author uses a manufacturing production metaphor to support his thesis. The basic requirements of production are to build and deliver products in response to the demands of the customer at a scheduled time, at an acceptable quality level, and at the lowest possible cost.

Production consists of tasks that are arranged in steps. The first thing we must do is determine the limiting step. The limiting step is the point where the highest leverage can be applied.

Leverage is the output generated by a specific type of work activity. An activity with high leverage will generate a high level of output. An activity with low leverage will generate a low level of output.

Managers and Leverage

A manager’s output is the output of his organization and the output of the neighbouring organizations under his influence. A distinction must be made between the activities that a manager performs and output that is actually achieved.

A manager’s output is the sum of the result of the individual activities having varying degrees of leverage. The key to high output means being aware of leverage.

Managerial productivity can be improved in three ways:

Increasing the rate at which activities are performed.
Incresing the leverage associated with activities.
Shifting from low leverage activities to high leverage activities.

High leverage activities can be organized in three ways:

When many people are affected by one manager.
When a person’s activity is affected over a long period of time by a manager’s brief, well-focused set of actions.
When a large group’s work is affected by an individual supplying a unique, key piece of knowledge.

Managers and Monitoring

How often a subordinate should be monitored depends on his task-relevant maturity. Task-relevant maturity is the amount of experience and prior performance with a specific task.

Monitoring is a form of quality assurance. Quality assurance should be added at the lowest-added-value stage of process based upon task-relevant maturity.

For example, for senior engineers with high amounts of experience, code can simply be reviewed at the time a pull request is created. Training and review before a task is started is required for junior engineers with low amounts of experience.

Managers and Calendars

As an individual contributor, I am used to Getting Things Done. This means getting things out of the calendar, email, Slack, and meetings and onto my TODO list that I sequentially grind through in a focused manner.

High output management requires the opposite mindset. The calendar must be used as a production planning tool by:

Filling holes between time-critical events with non-time-critical but necessary activities.
Saying no to work beyond your capacity.

Managers and Meetings

Manufacturers turn out standard products. Managers must take a similar approach pinning down the types of interruptions they see and preparing standard responses. This is the role of meetings.

Sets of similar chores should be batched together. These should be accumulated and handled at once in staff and one-on-one meetings. An event that was irregular must be standardized and turned into something regular.

There are two general types of meetings:

Process-oriented meetings take place on a regular schedule and are for knowledge exchange.
Mission-oriented meetings are one off events designed to solve a specific problem.

Managers and Decisions

The output of meetings are decisions. High output decisions need to settle six important questions:

What decision needs to be made?
When does it have to be made?
Who will decide?
Who will need to be consulted prior to making the decision?
Who will ratify or veto the decision?
Who will need to be informed of the decision?

Managers and Objectives

Difference analysis is determining what the environment demands of you today and what the environment will demand of you in the future. How you react to this gap is the key outcome of a planning process.

This is the basis of a successful Management by Objective (MBO) system:

Objective: Where do I want to go?
Key Results: How will I pace myself to see if I am getting there?

Managers and Dual Reporting

Mission-oriented organizations are split into business units with little tie-in to other business units. Function-oriented organizations are split into units grouped by function.

This yields Grove’s Law: All large organizations with a common purpose end up in a hybrid organizational form.

For managers to be high leverage:

They must accept hybrid organizations.
They must master dual reporting.

Managers and Modes of Control

There are three main ways that actions can be influenced:

Free-market forces
Contractual obligations
Cultural values

Management creates a common set of values through articulation and by example.

Given a set of conditions, there is always a most appropriate mode of control. This is governed by a person’s motivation and their environment. Environmental factors are an index of complexity, uncertainty, and ambiguity which is called the CUA factor.

Managers and Performance

There are two reasons why a person may not being doing their job:

They are not capable.
They are not motivated.

There are two ways to improve performance:

Training
Motivation

The ideal manager is an excellent coach that:

Takes no personal credit for the success of the team and as a result the team trusts him.
He is tough on the team and gets their best performance out of them.
A good coach was likely a good player at one time.

Managers and Feedback

The goal of a assessment is to improve performance. This can be done by assessing the employees skill level and closing gaps. And also by intensify the employees motivation.

There are three keys when delivering a review:

Leveling: clearly providing praise and criticism
Listening: employing all of your senses to make certain you’ve been heard
Leaving yourself out.

Given a performance problem, there are 5 stages in performance problem. The biggest step is assuming responsibility because it requires both an intellectual and emotional step.

Book Notes: Drive

Sun, 17 Feb 2019 00:00:00 +0000

Drive studies the science of motivation. The author takes the current carrot-and-stick model of motivation and expands it with three concepts. Autonomy is the desire to direct our own lives. Mastery is the urge to make progress and get better at something that matters. Purpose is the yearning to do something in the service of something larger than ourselves.

Motivation

The author separates motivation into three layers. Motivation 1.0 is the biological drive for food, water, and sex. Motivation 2.0 is built around external rewards and punishments. And Motivation 3.0 consists of Authority, Mastery, and Purpose.

The problem with Motivation 2.0 is that it may squash intrinsic motivation. It may also encourage unethical behaviour and short-term thinking. Motivation 2.0 may be useful for rote tasks. However, it is problematic for creative work.

Autonomy

Autonomy consists of four conditions:

Task autonomy (What a person does)
Time autonomy (When it is done)
Team autonomy (Whom they do it with)
Technique autonomy (How they do it)

Mastery

Engagement produces Mastery. Mastery begins with Mihaly Csikszentmihalyi’s concept of Flow. A task must match our abilities and not be too easy or too hard.

There are three rules to Mastery:

Growth mindset
Effort, grit, and deliberate practice
Asymptotic. Mastery is a journey, not a destination.

Purpose

Purpose is the desire to make a contribution to a cause that is greater than oneself that endures over the long term.

Purpose is expressed in three ways:

Goals that address profits and purpose
Language that expresses more than self-interest
Policies that allow the pursuit of purpose

Book Notes: 7 Powers

Fri, 15 Feb 2019 00:00:00 +0000

7 Powers analyzes the statics and dynamics of business strategy.

Statics is about “Being There”. What makes a business so durably valuable? Dynamics is about “Getting There”. What developments yielded this attractive state of affairs in the first place?

Power is defined as the set of conditions creating the potential for persistent differential returns.

The author lists the 7 Powers as:

Scale Economies
Network Economies
Counter-Positioning
Switching Costs
Branding
Cornered Resource
Process Power

Equations

The author defines a Fundamental Equation of Strategy as:

  Value = Mgsm
  
Where:
  M = current market size
  g = discounted market growth factor
  s = long term market share
  m = long term differential margin

Each of the chapters on the 7 Powers also derive a Surplus Leader Margin for each of the Powers. Surplus Leader Margin is the profit margin the business with Power can expect to achieve if pricing is such that its competitor’s profits are zero.

Surplus Leader Margin is an important concept to think through. However, I found the equations more academic than pragmatic.

1) Scale Economies

Scale Economies is a business in which per unit cost declines as production volume increases.

The benefit is that you can offer a lower price to customers over time than your competitors can. And you can decrease investment over time faster than your competition.

2) Network Economies

Network Economies is a business in which the value realized by a customer increases as the installed base increases.

Network Economies business are, in general, winner-take-all. This creates a massive barrier to entry from competitors.

3) Counter-Positioning

Counter-Positioning is when a newcomer adopts a new superior business model with the incumbent does not mimic due to damaging their existing business.

I found this chapter be the best written and the most insightful. The author lists multiple reasons why an incumbent may not be able to challenge an upstart:

Milking the old business or Negative Combined Net Present Value
History’s Slave or having a Cognitive Bias against the new business
Job Security or The Principal Agent Problem

4) Switching Costs

Switching Costs is when the value loss expected by a customer that would be incurred from switching to an alternate supplier for additional purchases.

There may be financial, procedural, or relational switching costs.

5) Branding

Branding is an asset that communicates information and evokes positive emotions in the customer, leading to an increased willingness to pay for the product. It is the durable attribution of higher value to an objectively identical offering that arises from historical information about the seller.

6) Cornered Resource

A Cornered Resource is preferential access at attractive terms to a coveted asset that can independently enhance value.

There are five tests to qualify as a Cornered Resource:

Idiosyncratic. If a firm repeatedly acquires coveted assets at attractive terms, then the proper strategy question is, “Why are they able to do this?”
Non-arbitraged. Price is not arbitraged away.
Transferable. This resource could create value at another firm.
Ongoing. Power remains over time.
Sufficient. Continued differential returns.

7) Process Power

Process Power is company organization and activity sets which enable lower costs and superior product, and which can be matched only by an extended commitment.

Process Power equals operational excellence, plus hysteresis.

Power Intensity, Benefits, and Barriers

Power Intensity can be outlined in a table:

The Benefits and Barriers to Power can be summarized in a diagram:

Power Progression

The author caps the book off with a description of Power Dynamics. How to get there from here.

At Origination, a firm must start with a Cornered Resource or Counter Positioning. Then at Takeoff, the firm must build Network Economies, Scale Economies, or Switching Costs. And when the firm reaches Stability, the firm must create Branding and Process Power.

Custom Xcode Templates for UIViewControllers

Sat, 29 Dec 2018 00:00:00 +0000

One way to enforce good software design and architecture is by using boiler plate code. Xcode supports custom file templates to help with this.

Let’s walk through creating a custom template for a UIViewController. First we will create the directory that Xcode looks for templates in. Then we will copy a template into this directory that we can edit.

$ mkdir -p ~/Library/Developer/Xcode/Templates/File\ Templates/Custom
$ cd ~/Library/Developer/Xcode/Templates/File\ Templates/Custom
$ cp /Applications/Xcode.app/Contents/Developer/Library/Xcode/Templates/File\ Templates/Source/Swift\ File.xctemplate Swift\ UIViewController.xctemplate

Next open up TemplateInfo.plist.

 $ open Swift\ UIViewController.xctemplate/TemplateInfo.plist

You will want to edit the Description and Summary fields.

Next open up ___FILEBASENAME__.swift.

$ open Swift\ UIViewController.xctemplate/___FILEBASENAME___.swift

I like to split up my UIViewControllers using extensions. IBOutlets, members, and existing initializer‘s must be in the class declaration. However, lifecycle code can be in an extension. I also put segues and actions in their own extensions.

//___FILEHEADER___

import UIKit

class ___FILEBASENAMEASIDENTIFIER___: UIViewController {

    // MARK: - IBOutlets

    // MARK: - Members

    // MARK: - Initializers
    
    required init?(coder aDecoder: NSCoder) {
        fatalError("init(coder:) has not been implemented")
    }
}

extension ___FILEBASENAMEASIDENTIFIER___ {
    
    // MARK: - View lifecycle

    override func viewDidLoad() {
        super.viewDidLoad()
    }
    
    override func viewWillAppear(_ animated: Bool) {
        super.viewWillAppear(animated)
    }

   override  func viewDidAppear(_ animated: Bool) {
        super.viewDidAppear(animated)
    }

   override  func viewWillDisappear(_ animated: Bool) {
        super.viewWillDisappear(animated)
    }

    override func viewDidDisappear(_ animated: Bool) {
        super.viewDidDisappear(animated)
    }
}

extension ___FILEBASENAMEASIDENTIFIER___ {
    
    // MARK: - Segues
    
    override func prepare(for segue: UIStoryboardSegue, sender: Any?) {
        // Inject dependencies in child viewcontrollers
    }
}

extension ___FILEBASENAMEASIDENTIFIER___ {
    // MARK: - IBActions
    
}

Now when you create a File -> New -> File, the Swift UIViewController template appears in the dialog box.

IBInspectable and UIButton Borders in Storyboards

Thu, 27 Dec 2018 00:00:00 +0000

iOS 7 ushered in the era of flat design. It also ushered in one major design flaw. Buttons do not look like buttons.

Text that is a button requires visual affordance that it can be tapped as opposed to text that is purely a label. The way that I style buttons is by adding a border with rounded corners. I feel this adds the required affordance while still staying true the minimal, flat design of the rest of the operating system.

I lay out views in Interface Builder. It is not possible to set the border out of the box. However, we can use @IBInspectable to help us.

// UIView+Inspectable.swift

extension UIView {
    @IBInspectable var cornerRadius: CGFloat {
        get {
            return layer.cornerRadius
        }
        set {
            layer.cornerRadius = newValue
            layer.masksToBounds = newValue > 0
        }
    }
    
    @IBInspectable var borderWidth: CGFloat {
        get {
            return layer.borderWidth
        }
        set {
            layer.borderWidth = newValue
        }
    }
    
    @IBInspectable var borderColor: UIColor? {
        get {
            guard let borderColor = layer.borderColor else {
                return nil
            }
            return UIColor(cgColor: borderColor)
        }
        set {
            layer.borderColor = newValue?.cgColor
        }
    }
}

After we add this file to our Xcode project we are able to set these values in the Attribute Inspector.

Book Notes: Thinking In Systems

Sun, 23 Dec 2018 00:00:00 +0000

Thinking In Systems analyzes the structures and behaviours of systems.

Donella Meadows was an environmental scientist, teacher, and writer. She primarly focused on systems thinking and how that applied to the environment. This type of thinking can also be applied to many other domains.

The beefy part of the book lays out a number of systems traps and analyzes the ways to get out of them.

Highlights

Page 1 · Location 189

The answer clearly lies within the Slinky itself. The hands that manipulate it suppress or release some behavior that is latent within the structure of the spring. That is a central insight of systems theory. Once we see the relationship between structure and behavior, we can begin to understand how systems work, what makes them produce poor results, and how to shift them into better behavior patterns.

Page 5 · Location 262

I start with the basics: the definition of a system and a dissection of its parts ( in a reductionist, unholistic way ). Then I put the parts back together to show how they interconnect to make the basic operating unit of a system: the feedback loop.

Page 5 · Location 264

Next I will introduce you to a systems zoo — a collection of some common and interesting types of systems. You’ll see how a few of these creatures behave and why and where they can be found. You’ll recognize them ; they’re all around you and even within you.

Page 5 · Location 266

With a few of the zoo “ animals ” — a set of specific examples — as a foundation, I’ll step back and talk about how and why systems work so beautifully and the reasons why they so often surprise and confound us.

Page 11 · Location 327

system * is an interconnected set of elements that is coherently organized in a way that achieves something. If you look at that definition closely for a minute, you can see that a system must consist of three kinds of things: elements, interconnections, and a function or purpose.

Page 12 · Location 346

A system is more than the sum of its parts. It may exhibit adaptive, dynamic, goal - seeking, self - preserving, and sometimes evolutionary behavior.

Page 14 · Location 384

Many of the interconnections in systems operate through the flow of information. Information holds systems together and plays a great role in determining how they operate.

Page 14 · Location 399

Purposes are deduced from behavior, not from rhetoric or stated goals.

Page 16 · Location 422

Keeping sub - purposes and overall system purposes in harmony is an essential function of successful systems.

Page 17 · Location 443

A change in purpose changes a system profoundly, even if every element and interconnection remains the same.

Page 18 · Location 465

A stock is the memory of the history of changing flows within the system.

Page 18 · Location 466

Stocks change over time through the actions of a flow. Flows are filling and draining, births and deaths, purchases and sales, growth and decay, deposits and withdrawals, successes and failures. A stock, then, is the present memory of the history of changing flows within the system.

Page 22 · Location 531

A stock can be increased by decreasing its outflow rate as well as by increasing its inflow rate. There’s more than one way to fill a bathtub!

Page 23 · Location 542

Stocks generally change slowly, even when the flows into or out of them change suddenly. Therefore, stocks act as delays or buffers or shock absorbers in systems.

Page 24 · Location 562

Stocks allow inflows and outflows to be decoupled and to be independent and temporarily out of balance with each other.

Page 27 · Location 613

A feedback loop is a closed chain of causal connections from a stock, through a set of decisions or rules or physical laws or actions that are dependent on the level of the stock, and back again through a flow to change the stock.

Page 28 · Location 639

This kind of stabilizing, goal - seeking, regulating loop is called a balancing feedback loop, so I put a B inside the loop in the diagram. Balancing feedback loops are goal - seeking or stability - seeking. Each tries to keep a stock at a given value or within a range of values. A balancing feedback loop opposes whatever direction of change is imposed on the system. If you push a stock too far up, a balancing loop will try to pull it back down. If you shove it too far down, a balancing loop will try to bring it back up.

Page 31 · Location 677

It is called a reinforcing feedback loop, and will be noted with an R in the diagrams. It generates more input to a stock the more that is already there ( and less input the less that is already there ). A reinforcing feedback loop enhances whatever direction of change is imposed on it.

Page 32 · Location 703

Reinforcing feedback loops are self - enhancing, leading to exponential growth or to runaway collapses over time. They are found whenever a stock has the capacity to reinforce or reproduce itself.

Page 39 · Location 805

The information delivered by a feedback loop — even nonphysical feedback — can only affect future behavior; it can’t deliver a signal fast enough to correct behavior that drove the current feedback. Even nonphysical information takes time to feedback into the system.

Page 40 · Location 820

A stock - maintaining balancing feedback loop must have its goal set appropriately to compensate for draining or inflowing processes that affect that stock. Otherwise, the feedback process will fall short of or exceed the target for the stock.

Page 48 · Location 924

QUESTIONS FOR TESTING THE VALUE OF A MODEL

Are the driving factors likely to unfold this way?
If they did, would the system react this way?
What is driving the driving factors?

Page 50 · Location 977

The central question of economic development is how to keep the reinforcing loop of capital accumulation from growing more slowly than the reinforcing loop of population growth — so that people are getting richer instead of poorer.

Page 55 · Location 1040

A delay in a balancing feedback loop makes a system likely to oscillate.

Page 57 · Location 1069

Delays are pervasive in systems, and they are strong determinants of behavior. Changing the length of a delay may ( or may not, depending on the type of delay and the relative lengths of other delays ) make a large change in the behavior of a system.

Page 57 · Location 1076

We can’t begin to understand the dynamic behavior of systems unless we know where and how long the delays are. And we are aware that some delays can be powerful policy levers. Lengthening or shortening them can produce major changes in the behavior of systems.

Page 59 · Location 1113

In physical, exponentially growing systems, there must be at least one reinforcing loop driving the growth and at least one balancing loop constraining the growth, because no physical system can grow forever in a finite environment.

Page 71 · Location 1267

Nonrenewable resources are stock - limited. The entire stock is available at once, and can be extracted at any rate ( limited mainly by extraction capital ). But since the stock is not renewed, the faster the extraction rate, the shorter the lifetime of the resource.

Page 71 · Location 1270

Renewable resources are flowlimited. They can support extraction or harvest indefinitely, but only at a finite flow rate equal to their regeneration rate. If they are extracted faster than they regenerate, they may eventually be driven below a critical threshold and become, for all practical purposes, nonrenewable.

Page 75 · Location 1313

Consider the properties of highly functional systems — machines or human communities or ecosystems — which are familiar to you. Chances are good that you may have observed one of three characteristics: resilience, self - organization, or hierarchy.

Page 76 · Location 1323

Resilience is a measure of a system’s ability to survive and persist within a variable environment. The opposite of resilience is brittleness or rigidity.

Page 76 · Location 1328

Even higher meta - meta - resilience comes from feedback loops that can learn, create, design, and evolve ever more complex restorative structures. Systems that can do this are self - organizing,

Page 81 · Location 1430

Systems often have the property of self - organization — the ability to structure themselves, to create new structure, to learn, diversify, and complexify. Even complex forms of self - organization may arise from relatively simple organizing rules — or may not.

Page 85 · Location 1500

When a subsystem’s goals dominate at the expense of the total system’s goals, the resulting behavior is called suboptimization.

Page 85 · Location 1512

Hierarchical systems evolve from the bottom up. The purpose of the upper layers of the hierarchy is to serve the purposes of the lower layers.

Page 86 · Location 1525

Everything we think we know about the world is a model.
Our models usually have a strong congruence with the world.
However, and conversely, our models fall far short of representing the world fully. That is why we make mistakes and why we are regularly surprised.

Page 87 · Location 1538

Everything we think we know about the world is a model. Our models do have a strong congruence with the world. Our models fall far short of representing the real world fully.

Page 89 · Location 1583

System structure is the source of system behavior. System behavior reveals itself as a series of events over time.

Page 90 · Location 1596

These behavior - based models are more useful than event - based ones, but they still have fundamental problems. First, they typically overemphasize system flows and underemphasize stocks.

Page 90 · Location 1601

Second, and more seriously, in trying to find statistical links that relate flows to each other, econometricians are searching for something that does not exist. There’s no reason to expect any flow to bear a stable relationship to any other flow. Flows go up and down, on and off, in all sorts of combinations, in response to stocks, not to other flows.

Page 91 · Location 1625

A linear relationship between two elements in a system can be drawn on a graph with a straight line. It’s a relationship with constant proportions.

Page 91 · Location 1628

A nonlinear relationship is one in which the cause does not produce a proportional effect. The relationship between cause and effect can only be drawn with curves or wiggles, not with a straight line.

Page 106 · Location 1920

Bounded rationality means that people make quite reasonable decisions based on the information they have. But they don’t have perfect information, especially about more distant parts of the system. Fishermen don’t know how many fish there are, much less how many fish will be caught by other fishermen that same day.

Page 107 · Location 1940

Economic theory as derived from Adam Smith assumes first that homo economicus acts with perfect optimality on complete information, and second that when many of the species homo economicus do that, their actions add up to the best possible outcome for everybody.

Page 111 · Location 2014

But some systems are more than surprising. They are perverse. These are the systems that are structured in ways that produce truly problematic behavior; they cause us great trouble. There are many forms of systems trouble, some of them unique, but many strikingly common. We call the system structures that produce such common patterns of problematic behavior archetypes. Some of the behaviors these archetypes manifest are addiction, drift to low performance, and escalation.

Page 116 · Location 2101

THE TRAP: POLICY RESISTANCE

When various actors try to pull a system stock toward various goals, the result can be policy resistance. Any new policy, especially if it’s effective, just pulls the stock farther from the goals of other actors and produces additional resistance, with a result that no one likes, but that everyone expends considerable effort in maintaining.

THE WAY OUT

Let go. Bring in all the actors and use the energy formerly expended on resistance to seek out mutually satisfactory ways for all goals to be realized — or redefinitions of larger and more important goals that everyone can pull toward together.

Page 119 · Location 2161

There are three ways to avoid the tragedy of the commons.

Educate and exhort. Help people to see the consequences of unrestrained use of the commons. Appeal to their morality. Persuade them to be temperate. Threaten transgressors with social disapproval or eternal hellfire.
Privatize the commons. Divide it up, so that each person reaps the consequences of his or her own actions. If some people lack the self - control to stay below the carrying capacity of their own private resource, those people will harm only themselves and not others.
Regulate the commons. Garrett Hardin calls this option, bluntly, “ mutual coercion, mutually agreed upon. ” Regulation can take many forms, from outright bans on certain behaviors to quotas, permits, taxes, incentives. To be effective, regulation must be enforced by policing and penalties.

Page 121 · Location 2202

THE TRAP: TRAGEDY OF THE COMMONS

When there is a commonly shared resource, every user benefits directly from its use, but shares the costs of its abuse with everyone else. Therefore, there is very weak feedback from the condition of the resource to the decisions of the resource users. The consequence is overuse of the resource, eroding it until it becomes unavailable to anyone.

THE WAY OUT:

Educate and exhort the users, so they understand the consequences of abusing the resource. And also restore or strengthen the missing feedback link, either by privatizing the resource so each user feels the direct consequences of its abuse or ( since many resources cannot be privatized ) by regulating the access of all users to the resource.

Page 123 · Location 2249

THE TRAP: DRIFT TO LOW PERFORMANCE

Allowing performance standards to be influenced by past performance, especially if there is a negative bias in perceiving past performance, sets up a reinforcing feedback loop of eroding goals that sets a system drifting toward low performance.

THE WAY OUT:

Keep performance standards absolute. Even better, let standards be enhanced by the best actual performances instead of being discouraged by the worst. Use the same structure to set up a drift toward high performance!

Page 126 · Location 2302

THE TRAP: ESCALATION

When the state of one stock is determined by trying to surpass the state of another stock — and vice versa — then there is a reinforcing feedback loop carrying the system into an arms race, a wealth race, a smear campaign, escalating loudness, escalating violence. The escalation is exponential and can lead to extremes surprisingly quickly. If nothing is done, the spiral will be stopped by someone’s collapse — because exponential growth cannot go on forever.

THE WAY OUT:

The best way out of this trap is to avoid getting in it. If caught in an escalating system, one can refuse to compete ( unilaterally disarm ), thereby interrupting the reinforcing loop. Or one can negotiate a new system with balancing loops to control the escalation.

Page 130 · Location 2382

THE TRAP: SUCCESS TO THE SUCCESSFUL

If the winners of a competition are systematically rewarded with the means to win again, a reinforcing feedback loop is created by which, if it is allowed to proceed uninhibited, the winners eventually take all, while the losers are eliminated.

THE WAY OUT:

Diversification, which allows those who are losing the competition to get out of that game and start another one; strict limitation on the fraction of the pie any one winner may win (antitrust laws); policies that level the playing field, removing some of the advantage of the strongest players or increasing the advantage of the weakest; policies that devise rewards for success that do not bias the next round of competition.

Page 135 · Location 2475

THE TRAP: SHIFTING THE BURDEN TO THE INTERVENOR

Shifting the burden, dependence, and addiction arise when a solution to a systemic problem reduces (or disguises) the symptoms, but does nothing to solve the underlying problem. Whether it is a substance that dulls one’s perception or a policy that hides the underlying trouble, the drug of choice interferes with the actions that could solve the real problem. If the intervention designed to correct the problem causes the self - maintaining capacity of the original system to atrophy or erode, then a destructive reinforcing feedback loop is set in motion. The system deteriorates; more and more of the solution is then required. The system will become more and more dependent on the intervention and less and less able to maintain its own desired state.

THE WAY OUT:

Again, the best way out of this trap is to avoid getting in. Beware of symptom - relieving or signal - denying policies or practices that don’t really address the problem. Take the focus off short - term relief and put it on long term restructuring.

Page 137 · Location 2524

THE TRAP: RULE BEATING

Rules to govern a system can lead to rule beating — perverse behavior that gives the appearance of obeying the rules or achieving the goals, but that actually distorts the system.

THE WAY OUT

Design, or redesign, rules to release creativity not in the direction of beating the rules, but in the direction of achieving the purpose of the rules.

Page 139 · Location 2566

GNP is a measure of throughput — flows of stuff made and purchased in a year — rather than capital stocks, the houses and cars and computers and stereos that are the source of real wealth and real pleasure.

Page 142 · Location 2584

THE TRAP: SEEKING THE WRONG GOAL

System behavior is particularly sensitive to the goals of feedback loops. If the goals — the indicators of satisfaction of the rules — are defined inaccurately or incompletely, the system may obediently work to produce a result that is not really intended or wanted.

THE WAY OUT

Specify indicators and goals that reflect the real welfare of the system. Be especially careful not to confuse effort with result or you will end up with a system that is producing effort, not result.

Page 164 · Location 2981

Systems modelers say that we change paradigms by building a model of the system, which takes us outside the system and forces us to see it whole.

Page 164 · Location 2983

There is yet one leverage point that is even higher than changing a paradigm. That is to keep oneself unattached in the arena of paradigms, to stay flexible, to realize that no paradigm is “ true, ” that every one, including the one that sweetly shapes your own worldview, is a tremendously limited understanding of an immense and amazing universe that is far beyond human comprehension. It is to “ get ” at a gut level the paradigm that there are paradigms, and to see that that itself is a paradigm, and to regard that whole realization as devastatingly funny.

Senior iOS Developer Checklist

Wed, 12 Dec 2018 00:00:00 +0000

I have been doing a lot of interviewing as I am helping Earnin build out a Vancouver office. I have had as many as three interviews on the busiest of days.

This has forced me to structure my evaluation and ranking of developers. I ask a number of cultural questions that are company and context specific. However, the technical topics are portable enough to make sense at most software companies.

I made myself a checklist of five main topics:

Programming Languages
Frameworks
Data Structures & Algorithms
Product
Process

Let’s expand on each bullet in the above list.

Programming Languages

The best developers have an excellent understanding of Swift, Objective-C, and C.

They have an understanding all the way from high-level language design to compiler implementation. They understand the basic syntax, control flow, and types. They also understand scope, memory model, and concurrency.

A senior developer can compare and contrast these languages at a superficial syntax level. And they can also describe these differences in terms of static vs dynamic, type theory, compilation, and runtime.

Swift

Basic syntax and semantics
Control flow
Strings
Collections
Functions and closures
Enumerations, structures, and classes
Memory model and lifecycle
Optionals and error cases
Protocols
Generics
Operators

Objective-C

Classes and objects
Methods and messages
Extensions
Protocols and delegates
Properties
Blocks
Errors
Collections
Memory and reference counting

C

Types, operators, and expressions
Control flow
Functions
Pointers and arrays
Structures
I/O
Header files
Preprocessor
Linker
Compiler

Frameworks

The senior developer has a broad understanding of many of the frameworks that are used day-to-day. But they can also compare and contrast them with 3rd party frameworks and frameworks from other languages and operating systems.

I am looking for understanding. But I am also looking for nuanced opinions on pros and cons.

Foundation
UIKit
Auto Layout
Core Data
Core Animation

Data Structures and Algorithms

This section will be controversial.

My belief is that you can be a productive member of a team as a junior or intermediate software developer without a deep understanding of computer science.

However, a senior developer must have a strong foundation of computing science theory, and even more importantly, be able to implement this in software.

That may involve implementing a cache, fixing concurrency errors, creating memory pools for optimization, or low-level integration with systems software.

This theory must also be engineered in a well-designed, well-tested manner.

Search
Sort
Lists
Associative Data Structures
Protocols: HTTP, TCP/IP, TLS, etc
Design Patterns
Unit Testing

Product

At least half of mobile development is interfacing software with the operator of that software.

I try to understand if the developers can critique and implement good design vs bad design. Do they have an interest and an understanding?

Do they think about the customer of the software when writing code? Do they think through business implications?

Design
Typography
Layout
Animation
Customers
Business

Process

Software is a team sport. Senior developers are able to take large feature requests and break them down into small TODO items that can be iteratively shipped.

They interact with product, design, QA to get this done.

They know how to use tools to effectively communicate.

Iteration
Source Code Control: git, Github, branching/merging, PRs
Bug tracking: JIRA, etc
Working with Design: Sketch, Zeplin, feedback
Working with Product: User stories, Job-To-Be-Done

Lessons Learned Shipping Impact for iOS

Sun, 09 Dec 2018 00:00:00 +0000

I shipped Impact earlier this year. Impact is an iOS App that helps CrossFit athletes plan and log their workouts. I’m happy with the progress I’ve made with it so far and after some reflection I have ideas where I want to take the App next. This is what I learned shipping version 1.0.

1) Ship Early

I shipped the first version of the App to TestFlight after a few weeks. This was very early–maybe even too early. The experience was buggy. The views and animations were janky. There was no signup or login flow. Saving and loading to disk or the network wasn’t even built. The data that had had just been entered would disappear every time the App was relaunched.

However, it was critical to ship this early. Was I on the right track? Did some bit of UI make sense on my screen size device but not on another? Should I prioritize one feature over another? Is adding an animation critical to creating a magic moment or is it just gratuitous? This feedback was critical.

Products consist of a few very large decisions and thousands of micro-decisions. Both are criticial to get correct.

It is key to let initial testers know that they are receiving early access to the App. The App shouldn’t be relied on to get day-to-day tasks done. It is important to let testers know that their feedback is important and that it will actually affect product development.

2) Ship Often

This leads to the next point. The best way to make progress is in short, bursty, iterations.

I shipped a TestFlight build almost every week. I paired this with a demo video that explained what was new in the build.

Sometimes, I would receive feedback from a build and put together a fix and ship another build the same day. This led to a positive feedback loop. It kept my morale up but it also kept testers engaged. They felt like their feedback was resulting in progress.

3) Listen to Athletes

Impact does not have users. Drug dealers have users. Impact is for CrossFit athletes. Impact is for CrossFit coaches. Impact is for CrossFit affiliate owners.

The way one speaks is the way one listens. If one calls people “users” then they will treat them like nameless, faceless slot machine pullers that should be served another ad.

It is clearer to map the job-to-be-done for a CrossFit athlete than a generic user. How does a CrossFit athlete prepare on a Sunday night for the week? How do they prepare before a WOD? How do they train during a WOD? And what about after? What language do they use?

How does an athlete interact with their coach? How does an athlete interact with other athletes?

Start asking questions. Lots and lots of questions. Ask what’s working. Ask what’s not working. Don’t ask for solutions–that’s your job. Ask about their day before they need to complete a task. Ask about their process. Listen for complaints about current tools that they use.

4) 50/50

I am a software programmer. My knee jerk reaction is to write more code. I guard against this by budgeting my time 50/50 between customer development and product development.

When in doubt, send more emails asking questions. Personally meet people when you can. Schedule FaceTime calls. At the end of each meeting ask for three specific people that you can speak to next.

Don’t take no for answer. Don’t take “I don’t know” for answer.

Ask clearly, “Introduce me to three people that you believe that an App like this could help”. Then shut your mouth. Let the awkward silence do the work.

This is how the flywheel of feedback spins up.

5) Don’t Make Work

The key learning from shipping 1.0 of Impact is that the App was too much work.

The CrossFit athlete archetype is a high-functioning, type-A personality. They wake up to train at 6am before work. Or they train at lunch time or right after work. They train 3 or 4 times per week or even more.

A CrossFit athlete spends $150 to $250 per month on their membership. They spend even more in food, supplements, and gear.

They work more than 40 hours each week. At least. They volunteer. They hike. They have a dog. They have other responsibilities.

What they don’t have is spare time for more work.

The fundamental learning from version 1.0 of Impact was that the App couldn’t make more work for the athlete without a huge payoff. The App should reduce work for the athlete in the best case scenario. And athletes are willing to trade money for that.

6) Make Magic Moments

The opposite of making work is making Magic Moments. The best Apps reduce the time to the first Magic Moment. There should be a Magic Moment during signup. There should be a Magic Moment every day. And there should be an even bigger Magic Moment every week.

Every feature should funnel athletes to the Magic Moment. Every tap and every scroll should be in service to the Magic Moment.

Magic Moments are sometimes called the job-to-be-done. Magic Moments are more than that. They fullfill both a functional job as well as an emotional job. They fullfill a need and also a want.

A Magic Moment is more than just getting the job done. A Magic Moment is getting the job done with style.