• Low-Bandwidth Collaboration

    Collaboration is low-bandwidth if the mediums in which groups of people coordinate require little resources.

    For example, face-to-face collaboration is high-bandwidth in that it requires parties to be in the same place at the same time with no latency between them and transmits information instantaneously using verbal and non-verbal (e.g. body language, whiteboards, etc) communication. Where as sharing memos asynchronously is low-bandwidth collaboration in that it is not tightly coupled with time (read it when you have time) and relies exclusively on a single medium (written communication) to transmit information.

    Both forms can be effective, but in different situations. Working on a distributed team necessitates low-bandwidth collaboration for processes to work smoothly, but things like design work is more efficient with high-bandwidth collaboration to get feedback quickly and be creative.


  • Learning Fringe Programming Languages Makes You Faster

    Learning a programming language that is immature and substantially different from what’s commonly used in the industry makes you faster at programming, specifically debugging and unblocking. These skills generalize to any programming and make you faster at ‘figuring things out’ next time something unexpected happens.

    For example, rust or clojure are significantly different from common languages like Java or Ruby that you can’t rely on previous experience to fumble your way through writing it (like JavaScript). Errors and tooling tend to be worse and often obscure (e.g. clojure’s error messages). This forces you to learn the language deeper to understand what’s going on and read much more code (e.g. reading source code of the language itself or a particular library).


  • Ego Depletion

    Baumeister and Tice introduced the concept that willpower is a finite resource that can be exhausted when used because it requires mental energy. The original experiments involved a willpower depleting task that requires self-control (e.g. don’t eat that cookie) followed by a task that requires a lot of mental effort (e.g. a quiz). They found that participants performed better when they were not first given the willpower depleting task.

    See also:


  • Web CLI

    A command line interface for a web application that seamlessly blends text based user input and graphical UI elements for output. This affords a highly productive keyboard-based interface (rather than point and click) which is familiar to developers.


  • Speed Is Undervalued

    Doing things fast has more primary and second-order benefits than we realize which makes it hard to conceptualize its full value. As a simple example, consider how we evaluate products by how completely they solve the problem. If you remove the temporal component, the value of the solution drops to zeroβ€”a solution is useless if it’s not available in a sensible amount of time.

    Now factor in the value of exceeding your expectation for a sensible amount of time. Then factor in the likelihood of doing it again because it’s fast. Then repurpose the time saved towards working on something else that is valuable.


  • Why There Aren't More Engineering Management Blogs

    The reason there are far fewer engineering management blogs compared to software development blogs not because there is a small supply of managers who can write blogs, but because managers are secretly worried they are doing it completely wrong (imposter syndrome).

    Unlike engineering blogs, managers don’t tend to write up how something went totally wrong with managing people lest they reveal too much or make themselves look bad.

    There’s also not a strong impetus to experiment. Experimental management sounds about as bad as experimental accounting.

    See also:


  • Hot Iron and Cold Iron

    Hot iron is rage and emotion. Cold iron is calculating and unwavering. When they face off together, cold iron tends to prevail.

    See also:

    • Malazan book of the fallen where thos expression comes from in discussing the difference between military leaders e.g. Coltaine and Dujek are cold iron where Korbolo Dom is hot iron.

  • Ellsberg Paradox

    People prefer situations where they know the risk. In experiments ran by Daniel Ellsberg, participants were asked to bet on a known 10% chance to win and an unknown chance to win (which was actually 90%). People tend to choose to bet on the 10% option.

    This is not considered risk aversion, but ambiguity aversion. People did not bet on the unknown option because it could be less than 10%.

    See also:


  • Stress and Anxiety Are Cumulative

    While it might seem like being stressed or anxious is binary, it’s more like a sum. Each stimulus accumulates even if they appear separate and unconnected e.g. personal vs work stress.

    The accumulation raises the temperature of your emotional thermometer and each additional stimulus feels larger than it is making it difficult to do anything. That’s why it’s surprising that something seemingly so small feels outrageously difficult or causes you to ‘snap’.


  • Signaling as a Service

    Signaling is often the underlying motivation for our behaviors. We make visible our values to both the in-group and out-groups in ways such as buying a luxury automobile (to signal wealth) or posting a flattering image of ourselves on vacation. Products and services can be viewed as selling signaling as the underlying value to users. For example Fortnite, which is free to play, sells costumes and emotes which have no affect the ability to win the game and are purely sold as a form of signaling.

    Read the blog post

    See also:


  • Robinhood Momentum Algorithm

    Robinhood is linked to recent events like the stock price of Hertz skyrocketing despite going bankrupt or Kodak jumping 1,000% on news of a pivot to drug manufacturing. By displaying stocks other users are buying/have bought (a simple way of consumerizing stock picking), they’ve inadvertently created a ‘momentum algorithm’ that, simply by displaying popularity more people buy and drive the price up.

    See also:

    • Money Stuff that suggests how COVID-19 has created millions of bored people looking for entertainment by betting on stocks

  • Identity Is a Powerful Motivator for Behaviors

    Behavioral change, such as forming a new habit, can be motivated by how it reflects your identity (both positively and negatively)–we do things that provide evidence for who we are.

    For example, identifying with being a healthy person means the actions you take will be more likely to reflect that. It’s not that you like to workout it’s that you are a fit person and that’s what fit people do.

    See also:


  • Synthesis Is a Fundamental Skill of Management

    A manager is constantly sythesizing information into useful direction and feedback to the team.

    For example, in performance management you are taking a body of work and extracting useful feedback. In team discussions and meetings, you are fascilitating discussions by making sure the important information is distilled and follow-up actions happen.

    Without the ability to synthesize vast amounts of disparate information the team can feel unfocused or unclear in what they need to do to be successful.

    See also:


  • PokΓ©mon Go Summer

    The platonic ideal of technology and human interaction occurred during the summer of 2016 when everyone was playing PokΓ©mon Go. As a location-based game, people were suddenly gathering in the same places with something in common to talk about and do together.


  • Operant Conditioning

    Learning through association of a behavior with rewards and punishment. We tend to do more of the things that make us feel good rather than feel bad.

    See also:

    • Atomic Habits which talks about designing for establishing good habits and stopping bad habits as a way of piggybacking off of operant conditioning (although I don’t think the term is ever mentioned in the book).

  • Gravity Model of Trade

    Trade flows can be predicted based on the proximity of two places and size of their respective economies. While not very detailed, it has shown to be very successful in it’s predictions that it is still used today, more than 50 years after it was introduced.

    Distance makes some intuitive sense, if trade costs anything, merchants would aim to reduce that cost. But, what underlies that is also potentially cultural similarities and language for doing business.

    See also:

    • A-Star algorithm which computes the shortest path between two points based on a cost function. (Could this be used along with the gravity equation as a predictor of trade routes?)

  • Fermi Paradox and Pessimism of Energy Economics

    A possible explanation to the Fermi Paradox can be found in energy economics. As a civilization progresses the cost of producing energy decreases which also reduces the cost of a world-ending event such as nuclear war. The pessimistic view is that civilizations eventually reach a point where they destroy themselves and therefore never make it to the point where they can achieve interplanetary communication or contact.

    See also: