Hack hack hack...

An open journal-- some of it written for you, but most of it is for me.

Linchpin by Seth Godin Notes

  • Notes:
    • Obedience is no longer rewarded - 4
    • We weren’t born to be cogs, we were trained to be cogs - 6
    • schools should teach: how to solve interesting problems, how to lead - 47
    • Emotional labor - 57
    • Jean baptist Colbert - 63: did he write down any of his thoughts?
    • Acknowledge fear, then proceeds - 65
    • Tom peters - 83
    • No one cares how hard you work- 95
    • Someone can do your job better and cheaper, not your art - 97
    • Artists are optimists- 98
    • Creativity is the instinct to produce - 102
    • Shipping -101-103
    • Steve McConnell - poorly timed thrashing 104
    • Elizabeth Gilbert Ted talk - 106
    • War of art, Steve pressfield- 107
    • Bio of lizard brain -109
    • You are a winner because you learn from losing - 115
    • 3 biological factors that drive job performance: social intelligence, fear response, and perception - public speaking brings together all 3 - 124
    • Fear is the most important emotion we have - it kept our ancestors alive. It dominates all other emotions. 124
    • Symptoms of lizard brain - 127
    • Leo babauta zen habits - 135
    • Shenpa: Tibetan word that roughly means scratching the itch - 138
    • Planing with the index card technique 146
    • Permission marketing? - 148
    • Pizzeria owner made shirts with bad Yelp reviews - 152
    • Damaging business book - Robert ringer - looking out for number one 153
    • Linus will never need to worry about finding work because he gave away Linux - 162
    • Sunny bates - 166
    • Clay shirky, Dan rushkoff - 169
    • Richard Branson - chartering a plane - 180
    • Can’t get attached to the objects of your attention -184
    • Art is defined by the fact I can’t tell you how to do it - 188
    • Your product should match your marketing, not the other way round -201
    • Deanna Vogt, Corp coach -202
    • Resilience, Ralph Waldo Emerson - 208
    • 5 elements of a personality: openness, conscientiousness, extraversion, agreeableness, and emotional stability - 210
    • Honest signals, book - 214
    • Dale Carnegie - 217
    • Creative -
    • Personal, original, unexpected, useful.
    • Unique creative - domain knowledge, position of trust, generosity to contribute back - 219
    • Org obey newton’s laws: a team at rest stays at rest - 221
    • Lester Wunderman: invented direct marketing - 222
    • Focus on making change down not up. Interacting with customers and employees is easier than bosses and investors - 229
    • Pranja - see the world objectively - 235

Patterns of Enterprise Application Architecture


“Architecture” is a term that lots of people try to define, with little agree- ment. There are two common elements: One is the highest-level breakdown of a system into its parts; the other, decisions that are hard to change. It’s also increasingly realized that there isn’t just one way to state a system’s architec- ture; rather, there are multiple architectures in a system, and the view of what is architecturally significant is one that can change over a system’s lifetime. p. 1.

The subjectivity comes in here as well because, if you find that something is easier to change than you once thought, then it’s no longer architectural. In the end architecture boils down to the important stuff—whatever that is. p. 2

Responsiveness is about how quickly the system acknowledges a request as opposed to processing it. This is important in many systems because users may become frustrated if a system has low responsiveness, even if its response time is good. If your system waits during the whole request, then your responsiveness and response time are the same. However, if you indicate that you’ve received the request before you complete, then your responsiveness is better. Providing a progress bar during a file copy improves the responsiveness of your user inter- face, even though it doesn’t improve response time. p. 7


There’s no generally accepted definition of a pattern, but perhaps the best place to start is Christopher Alexander, an inspiration for many pattern enthusi- asts: “Each pattern describes a problem which occurs over and over again in our environment, and then describes the core of the solution to that problem, in such a way that you can use this solution a million times over, without ever doing it the same way twice” [Alexander et al.]. Alexander is an architect, so he was talking about buildings, but the definition works pretty nicely for software as well. The focus of the pattern is a particular solution, one that’s both com- mon and effective in dealing with one or more recurring problems. Another way of looking at it is that a pattern is a chunk of advice and the art of creating patterns is to divide up many pieces of advice into relatively independent chunks so that you can refer to them and discuss them more or less separately. p. 9

Making a choice

Growing Object-Oriented Software, Guided by Tests

Object-Oriented Style

We value code that is easy to maintain over code that is easy to write. Implementing a feature in the most direct way can damage the maintainability of the system, for example by making the code difficult to understand or by introducing hidden dependencies between components. Balancing immediate and longer-term concerns is often tricky, but we’ve seen too many teams that can no longer deliver because their system is too brittle. p. 47

Encapsulation and Information Hiding We want to be careful with the distinction between “encapsulation” and “information hiding.” The terms are often used interchangeably but actually refer to two separate, and largely orthogonal, qualities: Encapsulation Ensures that the behavior of an object can only be affected through its API.It lets us control how much a change to one object will impact other parts of the system by ensuring that there are no unexpected dependencies between unrelated components.

Information hiding Conceals how an object implements its functionality behind the abstraction of its API. It lets us work with higher abstractions by ignoring lower-level details that are unrelated to the task at hand. p. 49

Our heuristic is that we should be able to describe what an object does without using any conjunctions (“and,” “or”). If we find ourselves adding clauses to the description, then the object probably should be broken up into collaborating objects, usually one for each clause. p. 52

Context Independence

While the “composite simpler than the sum of its parts” rule helps us decide whether an object hides enough information, the “context independence” rule helps us decide whether an object hides too much or hides the wrong information. p.54

Achieving Object-Oriented Design

Interface and Protocol Steve heard this useful distinction in a conference talk: an interface describes whether two components will fit together, while a protocol describes whether they will work together. p. 58

Kernighan and Plauger attribute the idea of pipes to Douglas McIlroy, who wrote a memo in 1964 suggesting the metaphor of data passing through a segmented garden hose. It’s currently available at http://plan9.bell-labs.com/who/dmr/mdmpipe.pdf. p. 66

The Sniper Makes a Bid

Defer Decisions There’s a technique we’ve used a couple of times now, which is to introduce a null implementation of a method (or even a type) to get us through the next step. This helps us focus on the immediate task without getting dragged into thinking about the next significant chunk of functionality. The null Auction, for example, allowed us to plug in a new relationship we’d discovered in a unit test without getting pulled into messaging issues. That, in turn, meant we could stop and think about the dependencies between our objects without the pressure of having a broken compilation. p. 136

When the facts change, I change my mind. What do you do, sir? —John Maynard Keynes p. 172

Working to Make Myself Obsolete

When I told my mother, a executive of 30 years mostly spent at Crabtree and Evelyn, about my goal to make myself obsolete she flipped.

“You are working yourself out your job,” she warned. “You need to be doing the opposite.”

My mom fought through sexism for decades to maintain her influential role at the top of the hierarchy. My mother is a scrapper and she clearly had to defend her territory. From her advice on management issues however, it seems pretty clear to me she probably never got ahead of the tsunami of work that consumes managers’ everyday lives. I mostly observe managers fire fighting and doing implementation. This is a trap. We all want to feel important, but if managers are the single point of failure, things will fall apart.

I’ve mused about my take on maker versus manager, it can be hard to let go, but those that never let go of the hybrid role are doomed. The manager may not be, but the team is. We owe it to our reports to make sure we have the time to properly manage them. It feeds our egos to have a full calendar and lots of “important” decisions to make, but if that’s all we do then we will eventually drown in our own self-importance rather than develop the people on our team to help shoulder the load. Scale will break us.

So thanks mom for the advice, but I still strive to make myself obsolete, so the world will continue to spin without me. So I can go on vacation without stressing. So I can dedicate a good chunk of my week to 1 on 1s and so my lieutenants feel qualified and empowered to make the calls shape our team and product. I may be working myself out of a job, but at least I feel like I’m going to do this one right.


Transparency begets trust and trust begets transparency. It isn’t easy and it feels unsafe to bare your process and sometimes soul. Talking about the thing you least want to talk about has been something that has helped our team address the weirdness that exists between our people and processes. But it wasn’t always like this. We used to ignore or only privately address what needed to be discussed the most.

A few months ago, during a period when a lot of new hires were starting and suffering from imposter syndrome, I tried to reinforce our culture of transparency and honesty by telling my team that I wouldn’t hire myself for our team. I assume that isn’t something that most managers admit and I explained my rationale. I’m not a strong enough coder to hang with them. I don’t complement the personalities we have. I don’t bring enough of a unique perspective as an individual contributor. It pains me to even write this, let alone tell my teammates that this is how I feel. Doing this led to a bunch of other team members describing their version of imposter syndrome and clearly put the junior members at ease. That is all to say, that saying the thing I least wanted to say helped build trust with the newest folks and let them know what I worry about which means they can be more sure when I’m confident. Always acting one way or another isn’t natural. Transparency, therefore, may start with the act of humanizing yourself and being vulnerable.

Team buy-in is tough to earn. It is much easier when the process is created and enforced by the team. As the manager, I can try to keep it on guardrails, but determining it and enforcing it myself would be a mistake. Maybe this isn’t sustainable with a bigger team. What if all the laws in this country were determined democratically by the entire population. It would be way too heavy. But with a team of 13, this is still possible and it is so much stronger when the policing is done by the team rather than me.

Transparency communicates trust. Exposing our vulnerabilities as an individual or an organization encourages candor and self-reflection. Putting a positive spin on everything doesn’t equate to boosted morale because it leads to whispers, eye rolling and eyebrow raising in the hallways. It is human nature to latch onto weaknesses, failures, and negatives and so if these deficiencies are not squashed publicly then they fester. They use rumors as sustenance and mutate in unpredictable ways. Talking about the hardest things right away is painful, but a preferable alternative to passively deferring to hearsay.

Transparency is hard. Generals must be able to make decisions. Soldiers must follow orders. We run into trouble is when soldiers don’t buy into the purpose or mission. Even the most ardent followers will lose enthusiasm over time if they don’t feel like they can influence the decision making process. And so what we find is people will do enough to keep their job. They half-heartedly go through the motions. In short order, that becomes the norm and apathy is a difficult virus to contain. And so while we can’t ask everyone for their input on every decision, if we know what is important to our people, we can ask them at the right times. This requires the decision maker to actually know their people and that takes a serious investment. Do you have time for that? I guess that’s up to you, but second guessing, rumor squashing, and low-morale also eats up time so maybe this would be worth some investment.

New Technology and Developer Happiness

What is the price worth paying to introduce a new technology into the stack? For our heavily junior team of 13 the price feels high. Our JS weapon of choice has been backbone and marionette. This toolset wasn’t determined by me. It was molded and implemented by a talented dev who might be a little short on leadership experience but has talent and intuition in spades. We’ve made some mistakes along the way, but the architectural choices he has made have served us well. Still about 8 months since its we push our first major feature set with marionette, the entire team has yet to be completely onboarded. We may be getting to the size where we can split our squad into front-end and back-end specialists, but to date that has never been discussed as a group. The fact that we all haven’t got there is a problem. It means that some of us aren’t capable to work on parts of the stack, which affects feature assignments and pairing.

This same dev is now suggesting that we introduce React to one aspect of the stack because of its rising popularity in the community. Our team spends a lot of time focusing on happiness at an individual level with 1 on 1s and quarterlies, but my priority isn’t individual happiness but rather team morale. At the moment, team morale is at an all-time high. Will introducing a new barrier to entry positively affect team morale because of its shiney? Will allowing the two JS leads on the team play with a new toy positively affect team morale? They both claimed in their last quarterly that their work brought them high levels of meaning and purpose and they found it challenging enough.

It is a tough call. I don’t have enough knowledge to know if this is truly a better tool or something new to learn for the sake of something new. How will we go about leveling up newcomers on two JS systems?

Goal Setting for Devs

I heard you, goals are important. I get it. I’ve watched the Ted talks and read the zillionth article on the importance of goals. I understand the psychology and the physiology. I’ve got a dirty secret though. I haven’t been able to set goals for developers. We’ve tried KPIs and they don’t seem to filter down to the individual contributor level. When I ran our apprenticeship program two years ago I tried weekly goals, bi-weekly goals, monthly goals and quarterly goals. The problem was, the constants, the areas the goals could be clearly defined were mostly areas of personal development – writing blog posts, learning keyboard shortcuts, giving a lunch and learn, etc. I had a much harder time defining goals for them to improve in their core job function, namely contributing well built features and pushing good code.

Recently we took a shot at changing our criteria for hiring and job responsibilities to being value based rather than bring task or milestone based. This rubric broke down our company values into behaviors and defined the expectations for each level within engineering. The resulting document sat well with the team and I think we are gettng closer, but I still can’t help but feel unfortable about the subjectivity of how to define a good productive dev. What devs do is complicated, which makes promotions and evaluation complicated. I don’t care whether you wrote a blog post this week if you pushed a great feature. I do care that you helped someone else push their feature or jumped in on a tough bug when everyone pretended they didn’t see it. But how do I formulate concrete actionable goals around that? Anything I come up with feels so arbitrary.

When we tried KPIs for the engineering team it made sense to have goals around the product. But for individual contributors who didn’t have a choice about what feature they build or how the product evolves from a high level, I could connect them to the department goal in a meta way, but not on an individual basis. How could they be held responsible for the adoption success a feature set, for example, that they didn’t have much agency in designing or implementing?

The best I can do for goal setting is to pull out actionable points from our longer-arching conversations and hold them accountable on the things that matter to them both. It feels like there has to be something better, but I haven’t been smart enough to figure it out.

Peer Mentorship

Next week I’m going to introduce a new 1 on 1 system on our team. I currently have 8 reports and the other manager on our team has 5. The load of weekly 1 on 1s is quite large if we are truly doing a legitimate check-in. And so, I’d like to try out a system of peer mentorship where peers conduct one on ones with each other for a week to 2 weeks a month. This means we will need to spend time training the team on how to conduct 1 on 1s which I will assume will force us to consider how we currently conduct our 1 on 1s and how they could be better.

Peer mentorship organically already exists on the team. We see natural gravitations along lines of technological specialities and personality types. I’ve done my best to find assignments for these pairs to share a chunk of time working together and I’ve always been impressed with the results. Formalizing this process through this peer mentorship check-in system should further reinforce that natural alliance without forcing devs to be people managers.

We’ll see how this experiment works out, but I’m excited to give it a try. Trusting team members with the task of developing their peers has almost always led to positive results. I expect the same here.

My Take on Maker Versus Manager

I toed the line for months trying to code a little and be the manager I wanted to be. When the team was at 8 I could find time to block off a morning without meetings or distractions. I could be a hybrid. As we grew to 12 it became untenable. The code I contributed was copy and CSS changes. Barely anything that was worthwhile from a contribution perspective. Letting go was the healthiest thing I could do. I’m not a coder anymore and I admitted that to my team in a frank conversation about imposter syndrome a few weeks ago. I’m a conductor. My job is to make sure the symphony continues to play in harmony. To draw out the sounds and rhythms of my individuals through 1 on 1s, quarterlies and conversations in the hallways.

Being an effective hybrid is possible. We have one. He is a great individual contributor and without fail he does his 1 on 1s and still pushes a ton of code. It works and he is getting the reps he needs to be a future CTO. Taking him off the keyboard would make him miserable and so he’ll toe the line for a while. He has 5 reports and I’m wary to give him more.

I’ve heard coders talk about how bad managers are out of touch because they don’t build anything anymore. I agree, which scares me. Which is why I want to try an experiment. I went on a paternity leave in October for a full month. I totally unplugged and trusted my lieutenants to carry the load. It went fine. No fires. Nobody quit. No month long rabbit holes anyone fell into. There was some maintenance and fine tuning to be done upon my return and some of the project assignments were off, but overall things went well. Meaning that if I disappeared for another month things would likely be fine again. Meaning I could become a dev for a month and let my lieutenants run the show again. I think it would be hard for me to resist my natural inclination to push, nudge and otherwise direct how the team runs, but what might happen if I stop going to all the high level meetings? We’ll find out. I’m going to propose we try it next month.

To be the manager my people deserve for a team of certain size I needed to let go of my drive to code. Sure I can nibble on small projects, but my days of plugging in and zoning out are seemingly over. Hybrids can be of value, but if you are running the show and managing anything larger than a team of 7-8 then you are doing yourself and your team a disservice by trying to focus on anything but your people.

Computer Science According to Spencer

What is it

Big O

  • used to classify algorithms by changes to the inputs
  • O(1) - constant time - kilometers to millimeters
  • O(log n) - opposite of exponential growth, searching binary tree
  • O(n) - iterating over an array
  • O(n log n) - good sorting algorithms
  • O(n2) - bubble sort (bad algoritms)
  • O(2n) - the chess board, example doubling something


  int rectangle_area(height, width) {
    int area = 0; // 1

    for(int i = 0; i < height; i++) { // 2*height + 1
      for(int j = 0; j < width; i++) { // 2*height*width + height
        area++; // height*width

    return area; // 1

  complexity: 3*height*width + 2*height + 3
  order: O(height*width)


  Constant time. O(1).

  int halves(n) {
    int count = 0; // 1

    if(n % 2 == 1) { // n
      n = n - 1; // n

    for(int i = n; i > 1; i = i / 2;) { // n + n + n
      count++; // n

    return count; // 1

  Complexity: 6n + 2
  Order O(log n)

  int bacteria_generation(days) {
    int total_population = 1; // 1
    int current_population = total_population; // 1

    for(int i = 0; i < days; i++) { // 1 + days - 1 + days - 1
      current_population = total_population; // days - 1

      for(int j = 0; j < current_population; j++) {

    return total_population;

  Order: O(2n)

  int max(array) {
    int max = array[0]; // 1

    for(int i = 1; i < array.length; i++) { //n + 1, [cost of determining array length](http://stackoverflow.com/questions/5872570/is-fixed-array-size-on-or-o1-in-space)
      if(array[i] > max) { // n
        max = array[i]; // potential n

    return max; // 1

  [3n + 3 complexity]


  int max(array) {
    int temp; // 1

    // sort the array in ascending order
    for(int i = 0; i < array.length; i++) { // n + 1
      for(int j = 0; j < array.length - 1; j++) { // (n + 1)(n + 1) -> n^2 + 2n + 2
        if( array[j] > array[j+1]) { // potentially n(n + 1) -> n^2 + 2n + 2
          temp = array[j+1]; // 1(n + 1)
          array[j+1] = array[j]; // potentially n(n + 1) -> n^2 + 2
          array[j] = temp; // 1(n + 1)

    return array[array.length-1]; // 1

  [3n^2 + 7n + 9 complexity]

  The first is more efficient.

  6. Which algorithm, foo or bar, is faster when when n is 10? 100? 1000?

  int foo(n) {
    int sum = 0

    for(int i = 0; i < n + 500; i++) { // 510, 600, 1500 operations

    return sum;

  foo(10) = 130,305
  foo(100) = 180,300
  foo(1000) = 1,125,750

  int bar(n) {
    int sum = 0

    for(int i = 0; i < n; i++) { // 10
      for(int j = 0; j < n; j++) { // 10
        sum++; // 1

    return sum;

  100, 10,000, 1,000,000

  bar is less work at 10, foo for the others