- Technological breakthroughs: what did they unlock that wasn't previously possible?
- Distributed databases: without an incentive, who would maintain it? Cryptoeconomics to the rescue.
- Monads: mostly related to solving side effects
- Bind: mostly for plumbing
- Set of
(m, unit, bind)is a monad (wheremis the "other" dimension of return, e.g. a debug string)
- IT cycles: expansion, conslidation, decentralization
- Value creation shifts upwards as old platforms become commoditized and standardized
- Current web distribution incumbents (e.g. Google, FB, Amazon, Apple) are too entrenched and too expensive to overcome
- Conslidating market; best to invest heavily into them and then progressively invest in new platforms that may grow to commoditize the incumbent business models
- Cryptonetworks: collapses the cost of building and scaling information networks by replacing centralized coordination with universal financial incentives
- Cryptoeconomic incentives incentivize independent 3rd parties to scale the network (leveraging the community)
- Diminishing capex of scale
- ==> Native business model of any network
- Cryptoeconomic incentives incentivize independent 3rd parties to scale the network (leveraging the community)
- Traditional valuation: price of stock reflects ability to monetize user base, rather than actual network value
- Key developer-side infrastructure (cryptocommodities) are still under-developed
- Networks don't die like businesses die; decentralized is inherently more robust
- Cryptoeconomic model is to a network what a business model is to a company
- Token valuations are more like national currencies: they track the overall activity within a network / economy
- Bonding curve: defines token price as a function of supply
- Allows us to have instant liquidity
- Dynamic inflation rates according to demand
- Difficulty in computing: every small adjustment of supply will change the price
- Sounds like a task for integration
- Child tokens have compounded risk, and are likely to be extremely volatile
- Power functions allow us to define our bonding curve in terms of a reserve ratio:
- ratio = poolBalance / (currentPrice * tokenSupply)
- poolBalance is the integral of the pricing function (e.g. 1/3*x^3 if pricing function is x^2)
- Often simplifies down to a constant value, e.g. 1/3 for quadratic curves
- ratio = poolBalance / (currentPrice * tokenSupply)
- Don't let managers make unilateral decision (Ray Dalio: believability weighted decision making)
- Give people freedom (Montessori method)
- Default to open
- Have an ambitious mission
- Create internal competition
- Create "bureaucracy busters" (aka fix what's annoying with your ops / processes)
- Culture isn't static; keep your values consistent
- Have a Chief Culture Officer (consistency over constancy)
- Invest more in hiring than in training
- Don't rely on inbound applications; seek out superstars
- When you find someone great, pursue them for years
- Only hire people who are better than you
- The CEO should review every hire
- If you're not a hell yes, you're a no
- Referrals can be a high quality source of hires: jog people's memories and go through their network
- Increasing hte referral bonus does not increase the referral rate
- Pay unfairly for top talent: human performance follows a power law distribution
- Be more like an all-star team than a family
- Most people aren't good interviewers
- Interview based on sample work and structured interview questions
- Find and develop your best interviewers
- Have at least two interviewers assess each value or key competency
- Create a safe space for employees to speak up: encourage revealing pain instead of concealing blame
- Encourage acting like an owner, rather than an employee
- People Ops should be transparent on the process and statistics, but private on individual details
- Constantly A/B test People Ops
- Remind people about bias, right before the performance review
- Identify and transition low performers
- Manager quality is the single best predictor of whether an employee will stay or leave
- Remind/nudge people frequently about how to be good managers
- Be a good coach
- Empower the team and don't micromanage
- Express interest/concern for team members' success and personal well being
- Be very productive/results oriented
- Help the team with career development
- Have a clear vision/strategy for the team
- Have important skills that help advise the team
- Survey employees about how well managers are doing on the above
- In training, spend more time on practice/repetition, and less on content
- Have the best employees in each area run trainings
- Give experiential awards, not monetary awards (non cash rewards trigger emotional responses)
- Reward thoughtful culture
- Hire only one-third "traditional HR" people in People Ops
- "The real company values, as opposed to the nice-sounding values, are shown by who gets rewarded, promoted, or let go"
- Behavious and skills we value in colleagues
- The best workplaces are those where you have stunning colleagues: direct all your processes towards attracting them
- "Adequate performance gets a generous severance package"
- If you're not going to fight hard to keep an employee, give them a severance package now
- Don't curtail freedom to avoid errors; relying too much on processes makes you fragile and lose out on talent
- Find responsible, self-disciplined people to avoid chaos when growing
- Increase talent density faster than complexity grows!
- Make mistakes, recover rapidly
- Processes:
- Good: help good people get more done
- Bad: attempt to prevent recoverable mistakes
- Avoid rule creep!
- Get outcomes through context, rather than control
- Inspire through goals, vision, strategy
- Highly aligned, loosely coupled teams
- Interactions are about strategy and goals, not tactics
- Compensation reviews should be adjusted to their market: "rehire" each year
- Individuals should manage their own career paths, rather than relying on their corporation for planning their careers
- Using complex math / logic to describe humans is often a mistake: using complicated measures for complex systems is systemically hiding risks / simplifications
- If you can't follow it, most will assume it's right (pose smart; appeal to authority)
- Principles:
- All value is perceived value (wealth is the cumulative perceived value of all parties); subjective value (vs. objective value)
- Controlled by each individual and can only be changed by that individual
- There is no unit of value
- An individual can only rank how they value things
- You cannot compare value judgements among mulitple people
- A price is only valid the instant it happened and only for the parties involved
- You cannot perform mathematical operations on prices to draw conclusions
- Only voluntary trade can reveal relative value of items
- All value is perceived value (wealth is the cumulative perceived value of all parties); subjective value (vs. objective value)
- Heuristic: when you get results that justify violating "do not do unto others what you do not want done to you", there was probably a mistake somewhere
- Challenge: a voluntary trade between two people may result in a loss of value to a third party
- However, the change in value of property due to 3rd party transactions is controlled by the owner, and only when a transaction is made with the property
- Modern neoclassical economics draws from microeconomic blocks:
- Utility functions
- Indifference analysis
- Kaldor-Hicks (cost-benefit)
- Austrian: reject all of these elements
- Utility functions vs. value scales
- Utility functions model individual preferences
- Value scales: ordered set of preferences
- Both are used to approximate utility maximization, either to maximize the function or finding the highest ranked feasible preferences in the set
- Austrians: scales are better because you can't measure the distance between preferences; utilities are not quantities and cannot be measured / have mathematical operations done to them
- Neoclassical: utility functions are not meant to be cardinal; they only signal preference; you can only monotonically scale (re-scale however you scaled it)
- Argues that Austrians got this part wrong: the math does not bound the thereoms to cardinal utility
- Indifference curves
- Utility of two choices are the same
- Austrian: however, action demonstrates preference; no preference can exist which cannot be revealed in action
- Neoclassical: there are more to actions than just the manifestation (what happened and what you can observe)
- Continuity
- Utility functions, supply and demand curves are continuous
- Austrian: human beings' perceptions are not continuous
- Neoclassical: without continuous functions, supply and demand will rarely be equal
- Welfare
- Austrian: every market transaction benefits all participants, every government intervention benefits some people at the cost of others
- Changes in social utility from intervention are ambiguous
- Neoclassical: using utility values allows you to make hard judgments about which choices are more or less efficient than others
- Austrian: every market transaction benefits all participants, every government intervention benefits some people at the cost of others
- Socialism
- Austrian: socialism is impossible because it lacks any method of economic calculation
- However, Austrian economics is all about qualitative judgments rather than quantiative judgments...
- The economic calculation problem is overstated as an argument against socialism; hardly empirical science backing it up
- Austrian: socialism is impossible because it lacks any method of economic calculation
- Monopoly Theory
- Neoclassical: free-market structures are "second-best" efficient; there is no feasible real-world way to improve them
- Austrian Business Cycle
- Widely accepted:
- Unemployment increases during downturns; caused by excessive wages and inflation as a unreliable means of reducing wages
- Widely accepted:
- Governance is in everything we do, the question is whether this governance is ad-hoc and implicit or formal and explicit
- Explicit governance is a protocol
- Both code and operators are governance questions in blockchains
- "Rules and regulations, laws and contracts, can never replace clarity of shared purpose and clear, deeply held principles about conduct in pursuit of that purpose." -- Dee Hock
- Can this clarity be encoded into a consitution?
- Argument for permissioned / known set of validator nodes: protecting decentralization and trust-minimization by using a binding constitution (that minimizes ability for parties to force actions from one another / control others)