Machines + Society #19: Value Exploration; solving weight loss; futarchy
A framework for systematic value discovery.
machines + society
Mako Shen | Feb 28, 2021
Creating a Value Exploration Syllabus
Fra Angelico, The Annunciation, c. 1438-47, (Convent of San Marco, Florence; photo: Steven Zucker, CC BY-NC-SA 2.0)
Back in 2017, I went to a conference where a kind and intelligent fellow by the name of Spencer Greenberg made a simple assertion: unless you know your intrinsic values, you are likely to fall into the trap of optimizing for the wrong thing. He gave the example of the investment banker trope who spends his 60+ hour work week career creating valuation models in Excel and slapping together Powerpoint presentations for his boss. The banker then looks back on his career with a boatload of cash and a similarly big boatload of regret and thinks, “I thought I wanted money when really all I wanted was to be recognized as excellent.”
Spencer explained that an intrinsic value was one that you valued for its own sake (e.g. being recognized as excellent), whereas an extrinsic value was one that only mattered because they led to some other outcome (e.g. gaining money). According to him, you want to discover and optimize for intrinsic, rather than extrinsic values. That is how you become satisfied and happy.
It took me a couple weeks of frustrated confusion to realize that although the intrinsic/extrinsic distinction sounded neat, it wasn’t all that helpful for me. Although I took Spencer’s online quiz and bought his general thesis— that better understanding of our values is key to flourishing— I was still unable to articulate my values. Worse yet, I did not know where to begin.
Two years later, I was talking to a notable monk who suggested I try to write down my own version of the Buddhist precepts. I tried, but it turned out to be an uninspired secular version of The Five Precepts. It kind of captured what I cared about, but frustratingly left a large majority of what I was often troubled by. I had values about romantic relationships, intellectual pursuits, and aesthetic preferences that I couldn’t articulate in the form of precepts. Once again, I put the effort on hold.
A couple months ago, I noticed a blog post about values by a fellow named Joe Edelman. He claimed that personal values were hard to pin down, but could be recognized by a number of specific characteristics; they were attention-directing (they tell you what aspects of a situation to focus on), improvisation-guiding (they guide action in new contexts), importance-recognizing (they take into account things that are important), and good life-constituting (they help you achieve a good life, whatever that may mean to you).
It wasn’t perfect, but it was by far the most lucid exposition on values I’d come across. I soon learned that he had a whole online textbook (Values-Based Social Design) dedicated to the exploration and articulation of personal values. This was exactly what I was looking for! Implicit in the writing is the notion that value exploration is complex, multilayered, and difficult, but certainly possible to do, especially with the . Also that a small group of dedicated people could make a lot of progress exploring these grounds together.
After a kind nudge from a friend, I decided to organize a small group around value exploration, taking what was most appealing from the textbook and adding what else I could think of. Below is the outline of the simple syllabus that I’ve created. I hope it inspires some of you to do something similar.
Values-Based Social Design: A Tentative Syllabus
This document attempts to create a broad plan for the systematic investigation of values in a group setting. It draws heavily on Joe Edelman’s Human Systems textbook.
Act 1: Building an internal map of values
Scene 1: Building a minimum viable ontology around values
Reading: See What The Hell are Values?
Here we discuss different ways to understand value, and closely consider some useful characteristics of values as attention directing, improvisation guiding, importance-recognizing, and good-life constituting.
This should grant us a simple vocabulary used to then think about values. Read the page, and see if each person in the group can explain one of the four aspects of a value to the others. Do you agree with the components? If not, what is missing?
Scene 2: Gathering evidence about values
Session 2 - Values from transitions
“A person's goals are evident in the plans they make and the actions they take. A person's preferences are evident when they choose one thing over another.
What's the evidence that we have certain values and not others? In this chapter, we look to our emotions, life transitions, and experiences of meaning (or meaninglessness) to gather precise information about values.”
Reading: Emotions, Values, and Wisdom
We will be thinking about different situations that have instantiated specific values (for instance, by using the transitions exercise). We will then think about emotions, values, and wisdom in the context of transitions.
(Session 3 - Values from other people
This session is similar to the last in theme, but instead of looking to critical periods within your own life, you will be interviewing someone else about a meaningful time in their life. Draw closely from the Value Excavation Teams exercise in the Human Systems textbook, but feel free to just use one person as both the interviewer and the values-checker.
Find a patient person to interview. It may take upwards up to an hour to even get one solid value down. Explain that the purpose of the interview is to explore a value that is important to them.
Ask the subject about a meaningful moment (five minutes to an hour, rather than months or years) in their life.
Clarify the context around the situation, asking specifically what they were paying attention to in the moment. From the exercise: “You are looking for an attentional policy—a way of approaching things that they find rewarding that relates to a context in their life.”
Pay close attention to any time they express a sentiment like “I was going into X looking for Y”. Write it down!
Clarify the value by checking that it is i) an attentional policy, ii) has diffuse benefits, and iii) provides a source of meaning. (For an explanation, see the exercise page).
After finding at least one value from another person, we will come together as a group and compare the values. What do they have in common? Were any especially surprising? Did you learn anything about trying to investigate others’ values?
Scene 3: Generating and dissecting values
Session 4 - Values Case Study: Integrity and Forgiveness
Reading: Notions of Integrity
In this session, we will look specifically at two values that are commonly discussed but which would benefit from more detailed elaboration.
What does integrity mean to you?
Begin by trying to describe a phrase that guides attention in a particular situation (e.g. “in a situation of doubt, paying attention to the difficult open questions”)
Describe some situations (perhaps from your own life, perhaps imaginary) where this notion of integrity is manifested.
Repeat the above for forgiveness. When is the last time you forgave yourself?
Session 5 - Writing Out Your Values
“Articulating a value in words doesn't necessarily help you live it better. But you still might want to write out a value you have for various reasons, including:
To tell someone what's important to you.
To find others who share the value, by circulating a text.
On a design project, to set a clear objective, or to check if users share the value, and whether your design helps them live by it.
To inspire strangers. If you phrase your value correctly, people who don't have that value yet might see the wisdom in your value when they read it. They will think to themselves "I should try being that way!"
Values are often discovered by: admiring someone, appreciating something in nature or human life, having difficult emotions (especially doubt, confusion, helplessness, shame, embarrassment, regret, grief) and realizing a new way you wanted to live or something that was newly important to you, or experimenting with how you try to act in a certain kind of situation.
In particular, make sure it's not a feeling or experience, a goal, or an internalized norm.”
Here we will attempt to write out in depth a few values that we care about. The emphasis is on clarity over quantity.
Scene 4: Acting out new values
Session 6 - Embodying new values
The aim of this session is to empirically test new values that you might want to embody but are unsure of. During the session, we will generate one or two values that you aspire to. Using tools from the previous sessions, write out in detail what that value consists of.
For instance, if you want to embody the value of vulnerability, or ‘when in a situation where I feel the possibility of being negatively judged, deciding to share anyways’, for this coming week, I will set a daily reminder on my phone to pay attention to situations when I can demonstrate the value of vulnerability.
At the end of the week, evaluate how you felt about embodying the value.
Act 2: Understanding values in complex social environments
Act 3: Designing values and systems around value
These last two sections are still a work in progress, but you can look to the next two sections of the Human Systems textbook to begin putting it together. Of course, if you have any suggestions for this syllabus, or end up trying it yourself, I would love to hear from you!
📰 Assorted Links 📰
What are the most important statistical ideas of the past 50 years? A statistics paper by the eminent Andrew Gelman and his colleague Aki Vehtari.
“We argue that the most important statistical ideas of the past half century are: counterfactual causal inference, bootstrapping and simulation-based inference, overparameterized models and regularization, multilevel models, generic computation algorithms, adaptive decision analysis, robust inference, and exploratory data analysis...”
It is interesting to realize that something so intuitive as cross validation was only so recently invented. Yet it only makes sense given that cross validation is only useful with sufficient computational resources.
An excellent list of questions and answers about laws governing speech and big tech platforms. On repealing Section 230 to prevent Big Tech deplatforming users:
Repealing Section 230 would do nothing to alleviate concerns about bias or censorship. As Senator Ron Wyden, one of the authors of Section 230, said, “I remind my colleagues that it is the First Amendment, not Section 230, that protects hate speech, and misinformation and lies, on- and offline. Pretending that repealing one law will solve our country’s problems is a fantasy.
Vitalik Buterin, one of the founders of Ethereum, on failures in prediction markets and the future of decentralized prediction markets.
Third, the demonstration that we saw of the prediction market working correctly will ease participants' fears. Users will see that the Augur oracle is capable of giving correct outputs even in very contentious situations (this time, there were two rounds of disputes, but the no side nevertheless cleanly won). People from outside the crypto space will see that the process works and be more inclined to participate. Perhaps even Nate Silver himself might get some DAI and use Augur, Omen, Polymarket and other markets to supplement his income in 2022 and beyond.
NFTs (non-fungible tokens) have been in the news lately. In brief, they are digital tokens for digital art that say something akin to “Elon Musk owns this limited edition of a NyanCat GIF <insert code for NyanCat GIF>”. As I understand it, the innovation is in inducing scarcity using techniques from cryptography. As a result, digital art creators are able to create a token that is scarce (since it cannot be copied), which means it can be traded. The Nyancat NFT was purchased a couple days ago for $600,000. My favorite NFT is this animated elastic heart bubble. My prediction is that NFTs will not really change how digital creative producers are compensated in the next five years. Unless a whole lot more people start regularly using cryptocurrency, the barrier to entry is far too high right now.
A very promising paper on semaglutide, a compound that significantly reduces weight loss (~30 lbs average in treatment group) with very few side effects. RCT, large sample size, plausible causal attribution — check. Here is the paper. Some authoritative comments. [H/t TC via Marginal Revolution]
The Clubhouse App and the Rise of Oral Psychodynamics from Zeynep Tufekci:
In oral psychodynamics, the conversational, formulaic styling dominates (which aides memory) as well as back-and-forth, redundancy, an emphasis on being less analytic and more aggregative, being more additive rather than developing complex and subordinate clauses (classic example is the Genesis which, like Homer’s Odyssey, is indeed an oral work which was later written down). Oral pschodynamics also tend to be more antagonistic, interpersonal and participatory.
Malware Analysis - Dissecting a PDF. There are surprisingly many ways a PDF can have malware embedded in it.
No Spoons (a.k.a. Soup Flight)
The entire menu always consists of 30 soups (most hot, some cold, most savory, a few sweet, most classic, a few experimental). They are only served in 1.5 ounce shot glasses, which you sip the soup from; no bowls or spoons. Anyone who brings their own spoon is permanently banned from ever coming again for the rest of their lives
Podcast of the month
The Past, Present, and Future of Personal Computing with David Smith. The most insightful podcast I have listened to about the past and future of the internet, from a living legend. David Smith created one of the first immersive video games and pioneered online collaboration technology.
🎧 Music 🎧
Meyem Aboulouafa - Ya Qalbi. Ambient electro ethereal Middle Eastern vocals.
Papooz - Ulysses and the Sea. Memorable chorus.
Indila - Dernière Danse (Ugg’A remix). For a dance.