Interesting Stuff from the Internet
Common knowledge, coordination, and strategic mentalizing in human social life | PNAS
The article "Empirical Analysis of the Gender Gap in Mathematics" investigates the global gender gap in mathematics achievement using data from international assessments like PISA. Key findings reveal that the gender gap varies significantly across countries and is influenced by cultural attitudes, societal norms, and educational practices. The gap is more pronounced at the higher end of performance, with boys more likely to be top performers. Stereotypes and biases about mathematical ability negatively impact girls' performance and interest. The study suggests that targeted interventions, such as addressing stereotypes and promoting gender-equal educational practices, can help reduce the gender gap.
Emergence - The Death of the Big 4: AI-Enabled Services Are Opening a Whole New Market (emcap.com)
Snippets from the Newsletters/ Newspapers/ Books
Utopia is on the horizon. I move two steps closer; it moves two steps further away. I walk another ten steps and the horizon runs ten steps further away. As much as I may walk, I'll never reach it. So what's the point of utopia? The point is this: to keep walking. - Eduardo Galeano
“When you repeat a lie, you spread it.
When you spread a lie, you strengthen it.
When you strengthen a lie, you become an accomplice to it.
In this disinformation age, we must do better.” - George Lakoff
Hawkins writes:
Intelligence is the ability of a system to learn a model of the world. However, the resulting model by itself is valueless, emotionless, and has no goals. Goals and values are provided by whatever system is using the model. It’s similar to how the explorers of the sixteenth through the twentieth centuries worked to create an accurate map of Earth. A ruthless military general might use the map to plan the best way to surround and murder an opposing army. A trader could use the exact same map to peacefully exchange goods. The map itself does not dictate these uses, nor does it impart any value to how it is used. It is just a map, neither murderous nor peaceful. Of course, maps vary in detail and in what they cover. Therefore, some maps might be better for war and others better for trade. But the desire to wage war or trade comes from the person using the map.
Similarly, the neocortex learns a model of the world, which by itself has no goals or values. The emotions that direct our behaviors are determined by the old brain. If one human’s old brain is aggressive, then it will use the model in the neocortex to better execute aggressive behavior. If another person’s old brain is benevolent, then it will use the model in the neocortex to better achieve its benevolent goals. As with maps, one person’s model of the world might be better suited for a particular set of aims, but the neocortex does not create the goals.
To the extent this is an analogy to AI, large language models are intelligent, but they do not have goals or values or drive. They are tools to be used by, well, anyone who is willing and able to take the initiative to use them.
Writer and scholar C.S. Lewis on why small choices matter:
"Good and evil both increase at compound interest. That is why the little decisions you and I make every day are of such infinite importance. The smallest good act today is the capture of a strategic point from which, a few months later, you may be able to go on to victories you never dreamed of. An apparently trivial indulgence in lust or anger today is the loss of a ridge or railway line or bridgehead from which the enemy may launch an attack otherwise impossible."
Sahil Bloom
a) The Law of Reversed Effort
There is a Zen parable that I absolutely love:
A martial arts student approaches his teacher and asks, "How long will it take me to master this craft?"
The teacher replies, "10 years."
The student, looking impatient, responds, "I want to master it faster
than that. I will work harder than anyone else. I will push myself to practice for many hours every single day. I won't rest until I become a master. How long will it take then?"
The teacher considers this new information, smiles, and answers, "20 years."
This story brings to life a concept called the Law of Reversed Effort, coined by author Aldous Huxley:
"The harder we try with the conscious will to do something, the less we shall succeed."
It's easy to find examples from your own life where this concept rings true:
When you press to try to complete a creative task, you become less creative.
When you actively push to try to find the perfect partner, you rarely find that person.
When you try to force yourself to fall asleep, you stare at the ceiling awake.
Elite sprinters follow the 85% rule: They try to run at 85% intensity because it keeps them loose, fluid, and effortless. When they try to run at 100% intensity, their body tenses up and they slow down.
The lesson here is simple: When you adopt a mindset of balanced effort, you achieve greater heights.
Life is not about pushing to the max at all times—this is a recipe for burnout and bad results.
Balance your effort, learn to breathe and flow in everything you do. If you find that headspace, you will always thrive.
The Clifford Principle: "It is wrong always, everywhere, and for anyone to believe anything on insufficient evidence."
Clifford argued that it is morally wrong to accept any belief without adequate evidence, as doing so promotes intellectual irresponsibility and can lead to harmful consequences. This principle underscores the importance of critical thinking and rigorous evaluation in decision-making processes.
18 of my favorite frameworks:
1) Location, vocation, relation: Where you live, what you do, who you're dating or married to. Try not to change more than one at a time.
2) Do something fast, do it for cheap, do it at a high quality. Pick two.
3) Head, heart, wallet: Three questions to ask yourself if you don't know what to write about — what does your head want you to write about? what does your heart want you to write about? what does your wallet want you to write about?
4) Frank Slootman's mantra for running companies: increase the tempo, raise the standards, narrow the focus.
5) High/Low self-monitors: High self-monitors adjust to the people around them. They're highly flexible and good at getting along with people, but they can be social chameleons. Low self-monitors don't adjust to their audience or change their behavior based on the situation. (Source: @GrahamDuncanNYC)
6) Voice, exit, and loyalty: If you're frustrated with a situation, there are three ways to respond: (1) voice means making a complaint or offering a suggestion, (2) exit means to leave the situation, and (3) loyalty means to put your head down and stay.
7) The shape of stories: A character who wants something encounters a problem before they can get it. At the peak of their despair, a guide steps into their lives, gives them a plan, and calls them to action. That action helps them avoid failure and ends in a success. (Source: Donald Miller)
8) At any point in a movie, you should be able to pause and ask: (1) what does the hero want? (2) who or what is stopping the hero from getting what they want? (3) What will the hero's life look like if they get what they want?
9) Focus on the 5-second moment of change to tell better stories: Every story revolves around a single, transformative moment In that moment. The clearer and more dramatic the flip or realization, the better the story.
10) One-way door vs. Two-way door decisions: One-way door decisions like selling your company or quitting your job are almost impossible to reverse, and you should make them slowly. Two-way door decisions like starting a hobby or changing your prices are easy to reverse, and you should make them faster. Most decisions are two-way doors.
11) The four types of work tasks: (1) need to do, love to do, (2) need to do, but don't love to do, (3) don't need to do, love to do, and (4) don't need to do, don't love to do.
12) The Eisenhower Matrix says there are four kinds of tasks: (1) Urgent and important, (2) urgent and not important, (3) not urgent and not important, and (4) important but not urgent. (See the photo below)
13) General and Specific: If you’re stuck on something general, zoom into the specifics. If you’re stuck on something specific, zoom out to the general. (From
14) Three rules for copywriting: Write things that are concrete, visual, and falsifiable. (From @harrydry)
15) Fast-twitch and slow-twitch thinkers: People exist on a spectrum. Fast-twitch people speak fast, work fast, and crave novelty. Throw a problem their way and they'll come up with 3–5 solutions in a matter of minutes. They're the people you want in the room when you're brainstorming an idea. Slow-twitch people are much more deliberate. Throw a problem at them and they'll ask for time to think about it, only to come back with something deeper, more polished, and more organized than a fast-twitch person could ever generate. You need both kinds of people for a group to be successful.
16) The TOP framework for finding what somebody should work on: Talent, organization, and passion. What are you talented at? What serves the organization? What are you passionate about? The perfect gig has all three.
17) The Keith Rabois formula for billion-dollar startups: "Find large highly fragmented industry w/ low NPS; vertically integrate a solution to simplify the value of the product."
18) The Sexy-Boring Matrix: When investing, look for companies in the unsexy + boring categories. Why? Because everything in the universe is a supply & demand curve. People want to work on things that are sexy and simple. Doing the opposite tends to lead to less competition, and therefore, higher returns. Without complexity, it's hard to be differentiated. And the sexy and complex categories attracts brilliant entrepreneurs who are hard to compete with (From Charles Songhurst).
a) My proposal has four components. At the core would be a wiki containing three major components: (a) an annotated and organized bibliography of published papers, (b) an organized set of ML experiment designs and analysis pipelines (with code), (c) an organized set of mathematical results and analysis techniques for proving results in optimization and learning theory.
b) This wiki would be maintained by the research community with the senior editors belonging to three panels: (a) subject expert panel of people who keep up-to-date within some research area, (b) methodology expert panel of experts in experiment design and statistical analysis, (c) mathematical panel of experts in the analysis of ML algorithms. Serving as a senior editor would be viewed as a prestigious career path with associated visibility and rewards. Senior editors would be well-placed to give keynotes and tutorials.
c) Authors would be expected to consult this wiki to identify relevant research results and methods. Failure to cite relevant work or methods that appear in the wiki would be grounds for rejection.
d) The second component: A submitted paper would include a structured abstract and a structured appendix. The abstract would state the motivation, research question, methods, and contributions.
e) The appendix would contain (a) An explanation for which wiki categories (and hence, prior work) they think are relevant and which ones are not relevant (but might appear to be relevant to a naive reader) (b) Pseudo-code of the experimental design and analysis pipeline (with access to the underlying code). A filled out checklist confirming that they followed best practices (i.e., they did not commit any of the common mistakes). (c) Statement and proof of all formal results. The purpose of this structure is the help the authors identify weaknesses prior to submitting the paper.
f) Component 3: Every paper is assigned an editor/mentor whose job is to help the authors improve the paper. The editor reads the paper and provides feedback on the "story": Are the motivation and claims well-supported by evidence? Is the research question interesting?
g) Will readers learn something? The editor checks that the appendix is complete (but does not check correctness of methods or proofs); the editor checks that the relevant prior research and experimental methods have been correctly identified and described. If necessary, the editor can post queries to a quora-like system where the subject experts, methodology experts, and math experts can post answers.
h) The editor sends their suggestions (mostly in the form of pointers into the wiki) to the authors. The editor may also determine that the paper is too flawed to be considered further (e.g., because the claims are known to be false or the authors are crackpots). After the authors have updated their paper in response to the editorial feedback, and the editor has approved it, the paper is released at this point onto a platform, such as @openreviewnet , that invites the research community to provide reviews and replications
I) Component 4: Papers can be nominated for a deep correctness check if they are regarded as important or surprising by the research community. An organization is funded (e.g., by national funding agencies) to pay reviewers to conduct these checks. The paper is sent to three kinds of reviewers. (a) People working on similar research questions: These reviewers evaluate whether the research question, claims, and evidence are accurately described. (b) Methodology experts (if the paper performs experiments): these reviewers assess the correctness of the methodology and analysis including auditing the experiments via selective replication. (c) Analytical experts (if the paper makes formal claims): these experts check the theoretical claims and proofs. The results of the check are published along with suggestions for follow-on research to address any problems that were uncovered.
j) The overall goal is an "edited @arxiv_org" with a second level of scrutiny for important papers. This should scale much better than the current system, which pretends but fails to carefully scrutinize everything. Equally important is that the community creates and maintains a wiki that captures the current state of knowledge in the field and gives authors better tools for "doing research right". Maybe this will solve the problem of the research explosion in ML?