Collective Knowledge #1: Cedric Chin - Tacit Knowledge, Mental Models and Accelerated Expertise

Summary

In this first episode, I speak with Cedric Chin, the founder of Commonplace, a blog about better business and career decision-making.

We explore ideas around learning when faced with uncertainty, the problem with frameworks, the classic rivalry between Generalists vs Specialists, scaling up approaches for learning digitally, how Cedric approaches knowledge management, and the possibility of developing explicit collective mental models.

0:01 - Introduction

8:57 - Learning when faced with uncertainty, the problem with frameworks

30:04 - Generalists vs Specialists: Understanding skill trees

47:12 - Scaling approaches to learning digitally

1:11:07 - Cedric's approach to knowledge management

1:23:03 - The possibility of collective mental models

Interview transcript

Evan Tan  0:02
Today I've got my first ever guest of the series, Cedric Chin, joining us. This is the collective knowledge series. The goal of this series is to discover and share the insights of professionals that are passionate about improving how they learn, and through the conversation, I hope to understand the guests' learning process, their tools and views on knowledge management, and how this shapes their worldviews.

And that's why I'm really excited to have Cedric here tonight. And really, we're going to try to address this fundamental question, how can we best organize knowledge individually and collectively?

To tell you a bit about Cedric, Cedric writes Commonplace, he's the founder of Commonplace, a blog that's about better business and career decision-making. It has a very active community over there. He was previously also the country manager, leading a team of engineers in a startup in Vietnam.

Cedric and I first crossed paths while I was working at a data analytics SaaS startup, I worked quite a fair bit with Cedric and I've always appreciated his thinking and the way he approaches things. I'm very excited to pick his brain today. So thank you for joining me, Cedric.

Maybe you can just share a little bit about about yourself and what you're working on right now?

Cedric Chin  1:23
Sure. So Commonplace is the name of the blog, it's at commoncog.com/blog. And the reason that it's at commoncog.com/blog is because Commongcog was the software that I was trying to build, and now I actually am starting to try to build. I started maybe about like, the last six months, or rather last quarter of last year. So that's when I started building it in earnest. And the blog originally started actually as a way to learn content marketing. And that's actually how Evan and I met, because I helped Holistics with content marketing for two and a half years, actually, after I got somewhat good at content marketing with the blog.

So the blog has done pretty well. And it actually now, roughly, I think, in the second half of the pandemic, in 2020, business mentor sort of like said, like, dude, like, I get a lot of value reading your blog, and you should figure out a way to monetize it, because then it gives you more ability, more options to do whatever you want on your entrepreneurship journey. And he was right.

And so I started the membership program in the end of 2020. It's grown since then. And basically, that covers most of my expenses. I mean, it's grown so much it's bizarre, if you told me that, like a blog would be able to cover all the company expenses and my living expenses, that it allows me to go and build software, I would say like, that's pretty ridiculous, because who ever heard of like, well, I guess there are some bloggers out there who can make money, right using their blog, but I've always regarded those as outliers. And I completely missed the wave of Substack writers. So there was a rise of Substack over the course of 2020, 2021, where independent content producers were able to charge subscription fees, and make a decent living just writing newsletters about topics that people find interesting. And Commonplace definitely falls into that category.

I think a lot of the, maybe half of the audience now are finance-oriented people, which is surprising to me. And then there's a good mix of tech people and programmers, data folk, but also like entrepreneurs, right. And the topics that I grapple with at Commonplace, it originally started as a career-focused blog. Because I had all these ideas about careers that I never really articulated. And a lot of my friends, when they were talking to me, they found it helpful. And when I say career, I mean, specifically, how do you navigate a career when you're not trying to climb a corporate ladder, for a big company?

I've always wanted to start a company. And so that meant like grappling with some of these ideas that I think most of my friends did not because they were more open to like working for a big company. And there the paths are more clear cut. But you tend to get lost in a field versus, you know, climbing a corporate ladder or like following the train tracks. And so some of the sort of ways of thinking about your career and navigating your career in this sort of free form way, I think I have better-developed ideas about those then like, sort of many of my friends and many people find that useful especially because like as anybody who has been around a bit in their careers, we know that careers are not that well structured. You do weird things you go you get bored of your job and then you switch industries. So you take some time off in a startup. I know I'm rambling a bit but let me sort of like bring it in.

So the blog started out like talking about career ideas, but eventually, it became a way for me to process a lot of my business experiences. When I was running the software engineering side of this Singaporean company, but the engineering office was in Vietnam. And over the course of my time there, we bootstrapped from basically $0 in product revenue, we were a consulting company, which means that we were over glorified, outsourcing shop, right, we build apps for clients. And we grew from nothing to $4.5 million over the course of like, two, two and a half years?

And this is like after one year of like, the very beginning was one year lost in the wilderness trying to figure out product. And so we successfully did that from consulting to product. And we, along the course of that journey, you know, I grappled with a lot of things that I didn't understand. And the blog became sort of this way to sort of process some of the things that we're seeing, like, I was dealing with a lot of like Chinese businessmen who are uneducated, which you see a lot in Southeast Asia, they were very good at business. Like they were ridiculously good at business. And they were, and they were superstitious, and they never had any formal education. And yet they were so good at business, and I couldn't figure it out. I was like, Why are you so good at business? What what is this thing, right?

And then there's also like, the limits of frameworks. So every time you read business books, or business blogs, or startup playbooks, right, go to market handbooks, you have this idea that like, this is the way that things get done, right in the startup world. And then you go and implement and you realize that, hey, it doesn't actually map, or like, our industry is like, it's weird. It's unique to us. Right? And in our particular case, the thing that really made a huge impact on me was, we made point-of-sale systems to mostly Singaporean businesses, right. And the point of sale business is, if you just do a cursory Google search, it's an incredibly competitive field, there are large companies, there are also a lot of like local companies selling to these local small SMEs, right? And when we talk to VCs by they will say, like, oh, you know, this is an incredibly crowded business, there's no way you're gonna make it big. There's no way you're gonna make a lot of money. And we were making a lot of money. So like, for I think, one year, when there's sort of that journey, right? When we, it was very clear, we were making in like a month, or we used to make in a quarter or like half a year, I can't remember what it was. But it was like, a ridiculous order of magnitude from consulting fees that we were charging, right?

And then my boss, and I were like, What are we missing? Why are we making so much money? It doesn't make sense. You know? And, and that's sort of like that experience, when we finally figured out like, what was causing it right. We eventually realized that the market was shaped in a certain way. And there was no amount of frameworks, no amount of like business books that could tell you like, this is actually what's going on, you have to go and go out to the uncertainty and, and grapple with it and seek out information. And if pieced together, this idea of like what's really going on in the industry.

So a lot of my blog now grapples with decision-making under uncertainty, which happens to be a topic that a lot of people, if you're in business, or you're in investing, you will eventually grapple with this. My blog tends to not appeal to people who are new to business, or novices. You need to have at least one or two experiences in your career, where you think the world is a certain structure, or you think frameworks are the best thing. And then you hit the limits of frameworks. And you realize that like, hey, no, like, there's more to this game than just taking frameworks and applying it. There are all these edge cases, if you have that experience, or you have dealt with like true uncertainty in your career in your life, you will be attracted to what I write about in Commonplace. And I think that helps explain why a lot of the audience recently has been more like hedge fund managers and finance people. Anyway, so that's the very long, incredibly convoluted explanation of like, why is it that I am writing a blog and maybe why we're talking about collective knowledge.

Learning when faced with uncertainty, the problem with frameworks

Evan Tan  8:57
Yeah, no, I think what you've just gone through there, that experience that informed how you're thinking about things, that was really informative for me to hear about that. Understand that, that's how it started this whole process for you. And that gives, there are just so many angles that we can look into right now, from what you just shared... Can you hear the air conditioning here? It's not too loud, is it? It's fine. Okay, perfect.

Because then that probably starts me off with my first question based on what you just shared. That whole situation where, basically, how do you think about learning in such conditions of uncertainty or, like you said, frameworks fail you? And basically, you're just thrown into a situation where you're trying to find some stable reference point or some certainty, but there, it's very hard to find that. How do you go about approaching this learning journey?

Cedric Chin  9:50
[Laughs] I'm laughing because I know you have first-hand experience with this because of our experience in Holistics together, right, like we were grappling with a market that we didn't really understand right. Or there are people that think they understand, but then there's always the local unique situation that your company finds itself in. And then there's always like quirks that really matter to you for execution, but maybe don't matter to like the people who give you advice like VCs or some analysts in Gartner, right?

So yeah, I grappled with that a lot. And to my surprise, I found that, or rather not to my surprise, I already suspected this somewhat. Frameworks are really the marker of a novice or a journeyman. I think journeyman is like the term where you're like an intermediate skill level, right? Not a master, not a senior journeyman, like a journeyman. So if you're a novice, or you're intermediate, intermediate skill level, you think in terms of frameworks. And but really, the frameworks are just ways for the masters to... well, there are like good frameworks and then there are shitty frameworks, like, probably some guy wants to have like a book publishing deal, so then he comes up with some frameworks in their business book, and then it goes viral.

But then there's also good frameworks by master practitioners and those frameworks, they're just the tip of the iceberg, right? The Master Practitioner cannot communicate the full thing that is in their head, that tells them what to do. Right. So they try to distill it into a framework, and then they try to communicate that to you. And then it's a very sort of like lousy compressed version of what actually lives in their heads. And so like, a smart operator, especially if you're dealing with somebody, you know that this is the fact of life, right, you have met a few master operators or master practitioners. And you can see that the way that they're operating and thinking about your specific problem, the way they're applying your expertise is not actually captured fully in the framework that they articulate to you, when you ask how they do what they do.

So then you begin to sort of like think like, hey, you know, there's something in their head that can't be easily articulated. That is the valuable thing. I know, I'm sort of talking around your original question, which is like, how do you deal with this pure uncertainty in the market when there's no framework, and I actually have a blog post that some people find quite powerful, which is like the reality of frameworks, right? It's like this idea of like, when you're looking at a situation first, don't try to apply a framework. Because when you apply a framework, it forces you into some sort of narrow tunnel vision, where you can, you're forced to sort of like... the framework shapes the perception, where you notice certain things prescribed by the framework, and you miss certain things right. And sometimes the framework might not be the right framework to look at. So you miss certain cues, because you're sort of blinkered by the framework.

So I have this thing where I say like, what I've learned is that you go in and you say, you don't know, even if you do think, you know, right, you sort of like I, I have no preconceptions, I'm just going to take a look at the entire environment as it is, and be completely overwhelmed with the information and the chaos, and the lack of patterns. And the reason why you have to be comfortable with that is that, the reason why frameworks are so popular is because frameworks really help to give meaning and shape to the chaos of reality. And so therefore, the novice tends to like sort of cling to it like a, like a lifeboat, right? Like a what do you call it, a life jacket?

And then after you sort of stew in this chaos, only then do you try to apply certain lenses. And so like, I find that the more masterful practitioners, even if they immediately know what to do they do some pattern matching, right, they are also able to sort of like take a step back and go like what's going on here? With far more ease and facility than a novice or an intermediate practitioner, who will sort of cling very tightly to the framework. It must be like this, it must be like this. It must be like this, because Geoffrey Moore wrote in Crossing the Chasm that it's like this, you know, and then you're like, Yeah, well, definitely Moore speaks from a certain context. And the ideas are very valuable and right, but the way they are expressed might be very different, maybe very, incredibly different, depending on your industry and your specific context. Right. So context does matter, right. So yes, like, sort of hold on to Geoffrey Moore's lens, but it's a lens, temporarily removing and take a look at all the data.

I'm being very abstract here I realize, which is maybe not that useful. But do you want to sort of go down this path more or like, are there any other questions, or maybe you want to go into a detailed example?

Evan Tan  14:29
I think, I think the interesting part of what you just shared, right, when you talk about how clinging to frameworks becomes a little limiting, because it's nearly like a sort of, if you will, confirmation bias of sorts. I have a framework, I have a lens, I'm going to look for things that fit into that worldview, you might be missing extra information. Okay, so if we talk about pattern matching and being able to, to understand, okay let's phrase it as... if you're trying to piece together a new field, you're trying to understand what's going on. You're trying to gather and collect information, sort knowledge out, prioritize things, organize and manage your collections of knowledge about something new. Surely there has to be something in how somebody can effectively do that, handle that, right? So of course, you don't have to be married to frameworks. But how do you? How do you actually become a better learner? How do you throw yourself into a new uncertain environment, look out for things that are important, pick things apart, figure out what, what do I need to know? To get in this field? Okay,

Cedric Chin  15:49
I see, I see where you're going. I'm so glad I asked you a follow up question than like going down my rabbit hole. I see where you, you're going with this. Um, I like to think that I'm better at this than many people. I mean, you'd be the judge of that, right? Like, you saw me sort of level up in marketing, you know, in our time together. So I will, I'll give you the answer that I gave somebody in Holistics. This was after like you left, right. But I only figured this out after you left and also sort of like trying to figure like, the marketing, sort of how to think about how to level up in marketing and apply that with the interface with sales. Because marketing and sales, like the interface tends to be a bit more complicated. I need to quickly figure out like, what is the state of the art in that particular field right?

So when I'm trying to do that, I think the key sort of like trick that you, that I use is, you want to get at the meta very quickly. And the meta is like, it's the meta game, right? Wait, so there's two pieces to this. One is you want to figure out what the metagame is, the meta game is like the edge of the field where like, you want to go where the experts are currently figuring out the limits, not because you can copy them, but because you want to know what the covered ground is. So you want to have a rough idea, like, okay, here's the current best-known practices around like how sales and marketing work together? Or like, here's the current best practices about doing content marketing, and it's changed from the last five years, five years ago, it was different. 10 years ago, it was different still, because like the environment changes, right? The competitive landscape changes, right? So that's like one half of what I do, I quickly go and find the best practitioners in the field that people sort of commonly say, are the best practitioners. And then I listen to them. And I try to piece together as I'm listening to them. And I specifically say Listen instead of blog posts, because, Listen, if you go find podcasts, sorry... So obviously here one precondition for this method is that the domain that you're trying to level up in has people who go on podcasts and give interviews. And the reason why podcasts are really important is because podcasts, during podcasts, especially when a practitioner speaks another practitioner, they are likely to drop, sort of like offhand comments that they have not rehearsed that reveal something about the field, right? Like, for example, you hear two marketing people talking and he said, Oh, you know, back in the day, like, it was so easy to play the Google game, the AdWords game was like, you throw in $1, you get $6 back, that kind of return it's never been seen ever since. And now the game is much harder. And so that, you know, in your head, like, okay, there's this cluster of experts who came up in that environment, who are used those kind of returns, and then now they had to adapt to that. And that's valuable knowledge. Because now you know, that's covered ground. Now you know, that maybe it's not a good idea to try to specialize in AdWords, right? Sorry, I'm recovering from COVID. So it's okay, it's fine.

So there's one element of it, which is like you go listen to them to figure out what where the edge of the meta is, and the other thing that I do is I go pick out landmark books and landmark books are books that have my definition of a landmark book is the conversation is like X right before the landmark book was published. And then after the landmark book was published, the practices and the conversation in that field change completely.

So an example of this [Evan: Triggers a paradigm shift of sorts?] Yes, correct. So an example of this in sales will be I think the most recent one was the Challenger Sale. Before the Challenger Sale, the last landmark book was... God, I can't remember what it's called. I think it's called SPIN Selling, SPIN selling. So all the sales books before SPIN Selling. We're talking from a sort of Zig Ziglar motivational, focus on the close sort of sale right. The tactics oriented around like sort of door-to-door salesman and insurance agents right, small dollars sum sales, and then SPIN Selling sort of like broke that paradigm and said dude guys, like, if you're doing enterprise sales, you're doing large five-figure dollar deals, right? The way you sell was completely different and SPIN Selling, you can tell from like the way people talked about it from, sort of like you follow the sales frameworks, they completely change after SPIN Selling came out, which was I think '90, I can't remember when exactly, it's either '87 or like '90, early '90s, probably '87, everything changed, and then like Challenger Sale, and then like, Predictable Revenue came out, I'd say maybe late 2000s. Again, I can't remember the date, I can probably pull it up, but it's in a blog post somewhere.

And then Predictable Revenue sort of like, was this Salesforce paradigm of like, basically figuring out a new way of doing outbound sales reach and the innovation there was sort of like spreading the people, the hunters from the what do you call it, I can't remember his term for it, it was like hunters and farmers, I think. And then after that was the Challenger Sale, right. So like, once you have sort of map out these like, okay, these are the landmark books, I can go read the landmark books, I know that every development in between is probably like small iterations, so I can ignore it. Or if you give me a new framework, I can like look at the data and go like, Okay, this is probably influenced by SPIN Selling, right? The other like, complex sales approaches that are, that are not worth mentioning. Because in my opinion, SPIN Selling sort of laid the foundation, they are just like tweaks, and you can very quickly figure out what the tweaks are and why.

So once you've got a map, right, and it's only like four or five books, usually for content marketing, again, it's like an equivalent set, I can give you an equivalent set. You read those, you map out the timeline in your head, you figure out immediately the names you go look for the podcasts, you listen to them. Now, many, many years after the original landmark books are published, they will usually drop hints, right? Like, oh, in a couple years, like I was talking to this guy, he forgot that this variant actually works well. And and now like most of Silicon Valley, use it and you're like, oh, I need to go check out this thing. And then you go and check on that thing. So that's figuring out the meta, and it's really quite a small investment in the grand scheme of things. I know a lot of people say like five books, holy shit, that's a lot. But if you think about it, you're taking five books that compress two decades right of experimentation, and found knowledge and covered ground into just like five books, however long it takes you to read five books. It happens that I read very quickly. I am quite a good reader, because I do a lot of it. So it's easy for me. But even if you're a bad reader, right, five books will take your five months, one book a month, right? That's five months in your entire career and immediately sort of like, okay, I know the covered ground. I'm better than, like, maybe 80% of the people in my field, who are, you know, that's learning whatever is being taught to them, or whatever they see on blog posts on LinkedIn, which is probably like some, you know, you don't even know whether it's like a fad or what it's built on, what the influences are. Anyway. So that's one half of the approach to the matter.

The other half of the approach is, I find that when people try to learn a new field, and this is especially true when it comes to marketing, alright. Because people, everybody always thinks that they know marketing. Marketing is easy. They've been marketed to their entire lives, they think they know how marketing works, especially if you're in sales, right? Sales is actually very, very similar to marketing, right? Because you need to understand the customer psyche, you need to be able to get under their skin, you need to, you know, figure out like what they're looking for. So it's like this element of psychology of product is also very similar to marketing. So people in product think they know how to do marketing, right? And so what I find very infuriating. And what I've learned not to do, is that when you come in with a background in sales or product, and you try to learn marketing, I'm using marketing as a specific example. Because this is like what I'm aware, very familiar with. Also, because it's like, it's a very pernicious example. You immediately try to map what the expert is saying, Whoever the expert is April Dunford, Seth Godin, you try to map it to something that you understand and, and there is a lot in marketing that maps something that you understand if your background is in sales, or if your background is a product, like empathy for the customer, the buyers journey, right? All these things map to the sales funnel, this maps to like sort of consumer psychology and on user psychology that you're taught to do in design interviews in product, right? And then you completely miss out on what makes good marketers, good marketers, because you're so busy trying to map it to some concept that you already have. You miss out on how the other like the actual marketer that you're learning from, sees the world.

And this is hugely wasteful, because I keep talking to people who are like, Oh, I understand marketing, okay, like, tell me when you understand. And then they tell me something about the buyers journey. And like on the surface level or the keywords they use are correct, right. But I can tell that they do not understand what it really implies based on their actions.

I'll give you an example. Right, like so a lot of people they think they understand the buyers journey, right? They don't understand the implications of the buyers journey. Like, if you're producing a piece of content where the funnel should be, right? What kind of content should be? Does it make sense for it to be there? Are there certain things that you shouldn't even try to put into marketing? Because it's not marketing's place because it's so far down the bias funnel? So the buyers journey, that it actually should be sales material instead? Right? So they don't, they can talk to you about like the buyers journey, right? And the psychological states that people in awareness, and what awareness, evaluation, all these these pieces, but then they don't have like this sort of deep internalized understanding of what this means in terms of like marketing tactics? Or like, where should we go after the funnel. And these are things that if you are, if you are so trapped in your existing sort of paradigms, about product or about sales, right? You will miss when the marketing person says these things, and then their implications, and there's a certain worldview communicated, because you're not listening carefully enough, you don't realize it actually, the way they look is subtly different from product.

And so what I did, in sorry, again, what I did in the months, this was after you left, I forced people working on the marketing team in Holistics, to explicitly say, are you wearing your marketing hat or your product hat right now, and you're not allowed to wear both. If you wear your marketing hat, you're not allowed to talk about product changes or product roadmap, your job is to communicate and sell and market the product as it is right now. Not what it's going to be six months from now. So when you listen to the customer, right, you're not allowed to think of new features. You're only allowed to think in terms of like psychology and copy and worldview and positioning. Right. You know, you're not allowed to think in terms of product words, which is like features and user journeys. And, you know, it's very easy to mix the two up because they're so similar, right?

And then like, and so like I will force people, like, you're wearing your product now, or you're wearing your marketing hat now? If you're wearing your marketing hat now, I don't want to hear about your onboarding flow. Right, because like, that's not our job and the worldview and frameworks, the way you think about it is completely different.

So as a result of this, I found that like, the acceleration of learning became much faster. Because they will now understand like, okay, if I'm listening to a talk by April Dunford, I need to put on my marketer hat, I need to pay very careful attention to the way that she approaches these ideas, or understanding, you know, customer psychology, understanding the ideal customer, that is different from the way that I as a product person, think about the ideal user, or the user psychology in the various states of the onboarding process. Right? Only then if I let go of my existing sort of like, understanding in the product field, can I actually internalize how these experts see the world? And that's a huge deal.

Evan Tan  28:23
Wow, that's a very powerful idea. It kind of reminds me of that quote, right? Is it, to a guy with a hammer every problem looks like a nail? Yeah.

Cedric Chin  28:33
I mean, no, that's the problem with frameworks, right? In general, I'll tell you a funny story. So I was talking to this guy called John Cutler, who is a product expert. He works for Amplitude. And his job is basically to go and talk to many product teams and help them with, help them become better product teams, because then you know, Amplitude wins, they win. And so he's seen hundreds of product teams. And he said to me, like one way that I can immediately tell if the person is an experienced product person or not, is if they talk in terms of frameworks. Because the really good product people, it's not that they don't do the things in the frameworks, they do them. But they don't talk about them. Like in terms of frameworks. It's just things that they do, right? Whereas the person who is a novice will like always call like, oh, this framework or that Northstar metric framework, that framework immediately, I'll be like, okay, like, you sweet summer child, how you don't actually know how to do product, you just sort of like glomming onto all these frameworks. So I was like, Holy shit, okay, that it's awesome. Like somebody else is telling me, it's confirming my biases against the frameworks or like, you know, frameworks at their place, but they're not the thing. They're the vehicle that the experts try to use to communicate the real thing that's in their head. And the quicker you understand that, the quicker you understand that the framework is just the captured thing that exists in your head and it's what's in your head that actually matters

Evan Tan  29:52
It's the articulation, it's crystallized, but it might not be the most current...

Cedric Chin  29:54
Yes, Yes. It might not be, yes it's not the full thing. It's a lousy compressed version of the thing. That's the only way that he can communicate it because words are limited.

Generalists vs Specialists: Understanding skill trees

Evan Tan  30:04
I think you're alluding to something you talked about in your on your blog about tacit knowledge, but I'm not going to go down that path yet. Yes, I think there is something here that was quite... that I would love to unpack. And it's going to be quite a big question. So approach it however you wish.

You brought up this point about surrendering your preconceptions in order to have that acceleration of learning, right? And then I have a question, for two kinds of people. One is a person that wants to get really good in their specific field, maybe it's marketing, maybe it's product management, maybe it's sales, versus somebody who needs to be more of a generalist, and in the situation described, a product person needing to understand marketing, or maybe a salesperson trying to understand product and all that.

Are those two scenarios very different, the person that's trying to go deep into the area, versus the person that's trying to get a broader feel of things. Are the ways of learning those two things completely, is there a different meta for both of them?

Cedric Chin  31:14
No, that's a really good question. Huh? So it's a really good question. Okay, so here, here, we get into somewhat speculative territory, I do have ideas about this, but they're not backed.

Like, like, the everything I just said earlier, I'm quite sure is true. Because I have not only verified it with other practitioners and experts, I've also verified it through my own personal practice. And this whole thing, like where whatever you believe is true, should should, should lead to effective action, right. So I can speak with some confidence over there, because I have already applied it in. And I've seen the results, and it seems to work. And I'm going to try keep applying it again. And again, until I find it to feel at which point I would probably have to update right. So I'm quite confident everything I just said in a previous sort of like, segment, um, this is the most speculated this is more. I'm not sure if this is true. But I will say this. So my hypothesis is that when you climb high enough in one skill tree, to be good enough, so So I think I think I'm pretty good at writing. Right? I think I'm much further I can vouch for that. I think I'm much further up in the writing skill tree than most people to the point where I can start, I can see what other writers have that I do not I can see the skill tree, I can see the shape of the skill. I know what sub sub skills exists, that if you're a novice or you're a lousy writer, you will not actually be able to sort of you will not be able to articulate much less notice that I can break down for you any any writer that I've read enough off, I can sort of break down what makes them there, right? Especially when he's able to do that not he or she is able to do them not right. And I can't tell you what's hard about what they do. And as a result of being this high up in one particular skill tree, because I spent so much so many years sort of like getting good at it. I have become more attuned to the skill trees in other skill trees. So I can tell you, for instance, right that I am nowhere near the stage of the skill tree where I can see the nuances and the shapes of other experts or other other good people's skill. That Let me paraphrase. Let me restate this. Um, you know how like when you're watching sport, if you don't play the sport, you don't you don't see the nuances and how you don't know how good they are like a person who doesn't play tennis. Right. You watch Federer and you're like, oh, he looks very nice, but you don't understand like, you know how bloody hard it is to play tennis the way Federer does a federal mix tennis love a belay. Right. Anyway, do you play tennis? I don't know. A little bit. Yeah. So So you probably have more of an appreciation for like what Nadal and Federer are able to do. That's so difficult, right? The nuances so the skill in a way that I do not in the same way that when I watch a judo match, I can I tell somewhat, but I know that I am not at the level of my skill in judo to be at the same level as my level of skill in writing. Because I know that I cannot tell the nuances of the skill yet off like say, Sure, he'll know like, why is he so undefeatable? I can't I don't understand that. I don't understand how he's so he makes fighting judo in finding all these world class opponents to lead playing with kids. I don't know why. Which tells me that I'm not at that skill level, yet in judo, to be able to To sort of like, disambiguate, like, Here are the various sub skills, okay? Now, that's a very valuable idea. The reason it's a very valuable idea is because it allows me to sort of like quickly, you know, like a sniff test, you know, when you wet your finger up in the air, like, like, how good are you actually at that particular sub skill. And recently, I am starting to feel I'm getting, I'm starting to get it when it comes to business, I'm starting to see things that novices don't in business, right. Similarly, in marketing, I'm starting to get a little right, in a way that very sadly, even though Judo is a passion of mine, I don't. So I'm able to sort of look at the market isn't go like, Okay, I think I can see the shape of their ability. And I can see all the various ways that I do not, I cannot do what they do. And therefore I know, in my head, the sub skills that need to go chase to get good at it. And I'll give you an example. Right? So like for marketing, there is this the shape of the skill in my head right now, and I could be wrong. Again, a expert marketer listening to this will probably find, like, various modifications, and I'm open to that, because I'm still, I still regard myself as like maybe a b minus in marketing.

There is a spectrum in marketing, where you are either very performance oriented, or you're very brand oriented. This is the word that that I think I think that phrase performance and brand, right? So and there's this tension in marketing and all marketing like false, even in like little things like, Should we have a pop up in our web page, there's, there's a tension there between brand and performance, right? So certain people are really good at performance. And these are very numbers driven. They, you know, they run ads, they can do a B testing, they figure out like, what's the ROI, the return on investment, or the return on on whatever dollar spending is, right. And in a small brand, where you sort of understand the consumer psychology and you're able to create campaigns like Estee Lauder, or I guess Sharpie is pretty good at this, I think they have a mix of both the really, really good marketers have both, right? And but then that's like a huge gradation of skill in both levels. And I'm actually not that aware of what exists in the performance marketing side. But in terms of like the brand side, at least on the content, I can sort of see certain things now that X, okay, this person is pretty good at breaking down, like what's necessary to create a consistent voice. And they know tactically and operationally how to achieve that, right. But then I can't do it yet. But I'm able to see it, which tells me that my skill is going to get there, right. So where I'm going with this is that this ability to sort of see the skill tree and the shape of the skill is very helpful, regardless of whether you want to quickly level up. Right, in the case of, in my case, I want to, I need to quickly level marketing. Because when I'm, when I left my previous company to go start my own entrepreneurship journey, I did a skill assessment, self assessment. And I was like, the way I'm most likely going to fail is that I'm not good at sales or marketing, I need to be good, at least one, right. And then I went off and did complacent. It's like a systematic sort of like multi year process to try to get to get good at marketing. So it's useful because then you quickly level out there. It's also useful if you want to climb because once you know the levels of the skill tree, at least, like for example, in writing, right, I know how to get good at certain things. The question is, like, do I want to put in the effort to do it? Because I know now, not only do I know, like the shape of the skill tree, but I also know like how much work it takes, and I know what the number of years it takes to get to that level. And then like I like, Okay, I'm going to work on this subsequent don't get me wrong, the sub skills that I'm going to if you ask me to describe this writing sub skills, they don't sound like you won't see them in blog posts. Or they're not easily articulated. All right, I was reading writers sort of analysis of program, right. I was like, this is not actually what makes program good. And the fact that the writer didn't know tells me something about that writer skill. Paul Graham's ability is really to distill an observation and observation down to like a very concise articulation. And during the articulation of that observation, he's able to use certain words that are very simple, but then when you put it together, they're like, they're like, beautiful. I guess like the wonderful example that would be like, relentlessly resourceful, right? Which is this wonderful articulation to us. They mean, like, they're very simple words, but they put together and then they express such huge vastness of meaning, that captured this observed property that if you're in startups, you know, when somebody is relentlessly resourceful, you can recognize it. Right? So that is the heart of program skill. It's this mix of like, observation and be able to articulate with like this wonderful, almost like when you look at the words, you're like, Yeah, this is the inevitable expression of this, this phenomenon that he's observed. Anyway, so I see I just needed to use like a whole number of sentences describe the sub skill. And this is sub skill that I think I can improve it, I know how to improve it, but it's gonna take years to be as good as Programming, right, but I can articulate what makes him good. That's different from my skill. And that is, I think one of the markers of like, if you're, you know, you're chasing expertise and you reach a certain level. Yeah. Does that answer your question?

Evan Tan  40:13
I hope. So. Let me let me see what I'm getting you right? Yeah, basically, are you suggesting here that by having gone deep in one area, one skill? Maybe it could be Judo could be writing to be someplace, I need to be better able to recognize, yes, what you need, yes, improve in another field? Which then leads me to, perhaps, a troubling notion that I'm trying to grapple with here, then, which is, how do you effectively transfer knowledge then? Because just now, you mentioned two ways, right? You said, there's usually a set of books that capture all the key ideas, compressed into maybe five books in a, in a certain area, say marketing and also podcast, that's one way they can really hear to practitioners trade and share ideas, and then really grasp the nuggets from what they discussed. But then, what does that mean for you know, the regular blog articles that you find online? That is his transfer of knowledge even possible in a in a effective way? Or is it that the people who have gone deep will get to know more and the people who are, are new to something they are going to suffer from the Dunning Kruger effect? They're gonna think they know something, but they really don't? Right?

Cedric Chin  41:35
I'm not, I realized that I should sort of wrap up my previous section with a more concise articulation of what I was trying to say. So what I was trying to say is yes, climbing high enough in one skill tree gives you the ability to recognize the state of skill in when when you're trying to learn a new skill. This is not a new observation. I think, I can't remember who it was. Maybe it's Peter Thiel, who like sort of said, like, if you're a chess grandmaster, you have like, sort of like, if you if you've excelled in one aspect of your life, it's likely you're gonna excel in other aspects of life. And he was probably referring something else like innate talent or intelligence or something like that. But

Evan Tan  42:19
the Shopify CEO said, I think, he'd be willing to hire Starcraft professionals. That notion...

Cedric Chin  42:26
yes. Oh, no. So sorry. So I the way that I'm coming at this is a bit more it's it's it's a bit it's a slightly different which is that I agree with with with the Shopify CEO, and when he said Tommy, Tommy Loki. But but the reason why and why I think this effect exists is slightly different than person intelligence or aptitude or like proof of hard work, which is that when I talk to people who have hit certain skill levels, like the minimum by quite like, the minimum bar, where like, beyond that you can see the shape of the scale. And then you know, like, maybe you choose not to go and chase it, because it takes too much effort. And you're interested in other things like, with me and writing, right? Because I seriously wanted to be a novelist when I was a kid growing up. But there is a point where you reach that and then you see, you see the full scale tree above you and you see all like the stars and like exactly why they're so good. And you know, like a rough idea of how what it takes to get there. There is this point, right. And every time I've talked to somebody who has reached this point, it could be a classical musician, who has spent like maybe seven, eight years practicing, it could be like a badminton player who stayed to play for the state. Right. But the standard I was a state player for Judo, but the standard for Judo is much lower than the standard for them. It's a new feature as you probably know, so, they are much higher on the skill tree right? Every time I talk to somebody like that, right what they know is that they know that the skill that there is a point beyond which you can see all of that. They also know that anybody who has above that point right, when they talk to each other, they say certain things that you do not understand. If you want proof. If you want an example of this right you go find this a YouTube video by Wyatt, where Jacob Kalia I think talks to Herbie Hancock and they are both jazz musicians. And the set of the video is like you know, I think Jacob Collier some extent explains music to a kid and then a teenager and then a college student of music and like a professional musician, and then Herbie Hancock which is like a multi Grammy Award winning musician, and Jacob Collier himself has won a Grammy Award right. When Jacob Kalia and Herbie Hancock start talking to each other. You do not understand what I'm talking about. Which tells you that you have not reached a level in the skill tree where you can when you can understand these two practitioners talking because if you have you would you should be able to understand what they say. They you do not understand what they're saying. You will not you can go and watch it. I think everything Jacob Collier makes sense right says makes sense until he hits Herbie Hancock and you're like holy shit. Why are you guys talking? About You know, they are talking about the the game, the real details of the game. And so when you know that that exists, that's Shiva, the skill tree exists, you learn to listen for it whenever you you hear two practitioners talking to each other, you learn to very carefully pick up Okay, wait, this is a very subtle quirk that a practitioner makes another practitioner, and then the other practitioner sort of laughs knowingly and it Kush over my head, right? Which tells me that I need to figure out what that is, I need to learn it, I need to get the point where I understand that. And so that's what I think. You have you, you learn and you internalize, once you've gone high enough in any skill tree. This is true whether you are high, like I know, a woman who is a coach for the woman's team and Ultimate Frisbee in Singapore, she sees this in the game of Frisbee. So she knows when she's like doing marketing that there is the same skill tree, it's always little it's levels to it. Right. And so she's watching out to it and, and I think I had a friend who told me that she had he had a new hire who was a classically trained musician for many years. And he switched to Tech, I did a boot camp. And he said that she seemed to level up and figure things out much faster than everyone else. And she zoomed in on like the higher levels of skill and paste in programming than all the other people that he met who had come from boot camps. And he suspected the transfer was going on. Transferring this very specific way.

Evan Tan  46:28
Wow. Okay, so it sounds like there's really no avoiding the hard work.

Cedric Chin  46:34
I think you can get good at listening. And hearing when a practitioner is making a, it's not very hard to notice that they're saying things you don't understand, right? And then like then making it your goal, your personal goal to say I'm going to go figure out how to get to the point where I understand what they're saying to each other. I think that's not too hard. Yeah, I think I really think that's not too hard. Once you you get an idea of like, what it sounds like you you very quickly learn to glom on to it. And this is where like marketing is particularly nice, because a lot of marketing podcasts are marketers talking to other marketers, because marketers like to talk, right? And they like to interview each other. Right? So that's beautiful. Yeah.

Scaling approaches to learning digitally

Evan Tan  47:20
I'll switch gears a little bit to some topic that's quite dear to my heart right now, which is trying to figure out how to take what you just described, and figure out if there's ways to scale that approach digitally. Because for me, I like to believe that, you know, there are a lot of people who are genuinely interested to better themselves and prove themselves, and they're scouring the internet for more information for ways to get better to improve. And what's your take on? How can we? How can we facilitate these transfers to happen more quickly at scale, to help people to get to that state of hitting that point in the skill tree where they can see the rest of this map of concepts and ideas? Or what's necessary to get better? Can we accelerate that process, that skill, for people?

Cedric Chin  48:22
So, um, I, I think, so I can address this question with like, there's a there's actually a huge branch of the Human Factors research crowd, mostly funded by the military that specializes in exactly this question. But it's not. It's not really technologically scalable, because a lot of their methods are, I mean, it is technologically scalable. But first, you need to learn the methods of how to extract expertise from the hits of people, right. And he doesn't actually get to the what I just described to you about the whole skill tree thing. So that the skill trading is actually something that I have not written about, because I... I don't know how to articulate it. I don't know if it's true. I've loved anecdotal evidence that it's true from people who are experts. But I don't actually know confidently that it is something that exists, right? I could give you so I will say this, anybody who's listening to this, you should go to my blog, and you should go to the passive knowledge series and look into it. And the reason I'm saying that's because I'm not going to go into it now. Because I don't think this is what you want, right? It's not scalable, scalable, because it requires you to learn how to learn there are methods for extracting tacit knowledge. And then you need to learn your methods for designing training programs, which are scalable, but then like learning these two things. It's not really a technology thing where it's like, you need to learn it and it probably would take one or two years of your life, and I happen to be interested in it. So I am trying to learn it. But it's not a scalable thing, right, and the output of that process Like once I learned that I should be able to design training programs and simulation programs and digital tutors which they have done for the military. Right? That will be scalable. And that will allow like the acceleration of expertise. But that's probably not what you're asking for. So those of you who are interested, you should go to my blog and read that series, tacit knowledge series common cog, type that into Google, you should find it. So I think the direction that I will go into that I think will be more useful to you and helpful to people watching this. At least in the context of like, what you're doing with centrally Did I pronounce that correctly? Centrally? Yes. That's right. Yeah. So the you seem the most recent blog post about note taking?

Evan Tan  50:41
Oh your note taking? Yeah, briefly, I haven't gone through.

Cedric Chin  50:47
I think that's, I think that's a lot more applicable to what you're doing and a lot more scalable. So I will articulate the theory is that this will take a bit of time, because it's a really big theory. And I'm still sort of digging into the implementations of the theory. So how I got into this is, there's this book called Accelerate expertise, which is a summary of the best, the best known techniques for accelerating expertise commissioned by the US Department of Defense, and written by a whole bunch of like the who's who have expertise, researchers, including one of the founders of the entire field, Paul felt vich, who was a contemporary of Erickson, and helped create the deliberate practice paradigm. Right. So that book, incredible book, incredibly dense book, incredibly difficult to read. At the very end of the book, they say, look, every accelerated training program that we know of, we have helped create, actually sort of sits on top of two learning theories, cognitive transformation theory, and cognitive flexibility theory. And these theories are not theories that you normally hear off, because these theories are not theories that are actually applicable to the real world. I mean, sorry, not applicable to classroom research. So a lot of the research that you read online, they're actually optimized for classroom learning environments, which is quite useless to me, if we are in a real world operating environment, or we are practitioner trying to get good at our careers, and some skill that is useful to business, right, or vocational education, if you will. A lot of research that is done and therefore have a lot of has a lot of attention tends to be optimized for test taking, or like student performance on tests, which we don't really care about, and the military does not care about. The military does not give a shit about like how well you do on test, the military cares about you doing well on the battlefield, not dying. So the theories are a bit unusual. And they're worth looking into. We're the theory that I want to talk about today with regard to this question, how to scale this is called cognitive flexibility theory, which ironically, was originally developed, sorry, for more mainstream education, but it was built out of work, understanding and trying to accelerate medical expertise. So think, the level of the doctor's education with the doctors in the hospital treating actual patients, not the level where they're still sort of like medical students in electric data, right. The other theory, cognitive transformation theory is also very interesting. It's about how real people, how people learn from real world experiences. And that theory, to sort of whet your appetite. But we're not gonna talk about it explains why some people can become masters to trial and error, and some people cannot. So we're not gonna talk about that we have our cognitive flexibility theory CFT. So CFT is about accelerating. It was built of work on accelerating medical expertise. And there are four big ideas that we need to talk about. So yes, two major claims. But before we get to the major claims, we need to talk about two big ideas that you need to talk about, we need to articulate how to set up for the claims. So I'll go in order. The first idea, the first claim that CFT makes so CFT is a learning theory, a theory about adaptive expertise, which is a very specific subfield of expertise research that asks how our expert experts able to deal with novelty. Because that is one of the elements of expertise. If you go back to drivers and drivers, like that's one of the articulations of like, why expertise is in addition to pattern matching and doing well they're able to deal with novelty. And all the classical theories of expertise research, do not deal with this question, right? Like, if you look at the chunking or theories by the group, when they look at chess players, where is it at chess players see chunks, and then when they see a certain board format, they're able to unfold the chunks to figure out like what needs to be done.

That that is that explains pattern matching parts of expertise, but it doesn't explain handling novelty. So CFT is a theory about novelty. The first big idea is that CFT grapples with ill-structured domains. It's specifically about structure domains. an ill-structured domain is a domain where there are concepts, right? But the way those concepts show up in practice are incredibly variable. Medicine is an ill structured domain. So medicine has concepts like a heart attack, right heart attack, some you can learn the textbook with a mechanical infection. But if you are a doctor, and you're trying to diagnose a heart attack, right, knowing the concept in a textbook is quite useless to you. Because the way the heart attack shows up is incredibly variable. It depends on the age of the patient, the complications that the patient has the previous diseases, the patient is the race of the patient agenda, the patient, right? And he can show me in a gajillion different ways, right, like, okay, not a good Jillian, but like many different ways. So like, we think of heart attacks, like ah, and then you fall, right. But somehow the attack started presenting as indigestion. They look like indigestion at first, and someone who attacks can last days, right. So it's not easy to sort of like go like, Oh, here's a heart attack. Let me teach you the concept in a textbook, and then you can do it. So it's a URL structure to the formal definition of structured domain is like, there are concepts for the way concepts are instantiated are hugely variable for the same nominal type. And, you know, this should sound somewhat familiar familiar to you, because businesses like this right? Business is 100% structured the meaning you may have competitive advantages, right, like bran, or scale advantages, but the way it looks is dependent on the industry, dependent on the competitive environment in which a competitive advantage was created dependent on a huge number of things. Right. Okay. So like, that's big idea. Number one, we are talking about infrastructure to means right, which is not School for the most part. Big Idea Number two, when you're dealing with ill structured domains, right? Cases are as important if not more important than concepts. And why is this the case, right, so like, this is actually quite a subtle argument. And so like, we need to talk a bit about about it a bit more. I think in school, we are taught to that concepts are more important than the cases. So the example I like to give is solving quadratic equations in secondary school in math, we are taught the technique to solve the quadratic equation, we're given one or two examples, and then we throw it away, right? We know that examples are important. It's like the steps the formula to solve the quadratic equations is important. And we internalize this. And I find that people who either are from stem backgrounds, there are software engineers, like me, or they're engineers, or they have gone through management consulting, they tend to think like this as well, the concepts are more important. The cases are truly, but if you are ill structured to me, right? What is theory? What is researchers found is that like good experts in instruction, that means they're able to, they don't reason from the concept, or the principles or the first principles, right, because it's too variable. What they do is they think, in terms of like what cases they've seen any combined fragments from cases that I've seen, to be able to reason about, like what they're looking at. Because you cannot, the backing for this research is that they find that junior doctors cannot go from concept or mechanism of disease, to what you quit symptoms in the real world, they also cannot go from observe symptoms all the way back up to like the concept. It's just too difficult. It's too variable, right. And then they find that expert doctors don't do that an expert is actually just reason from fragments of previous cases that they've seen. So this brings us to the two main concepts that we the two main claims of the theory, the two main claims is that experts in real structure domains are able to do what they're able to do. Because, first of all, the reason from my analogy to previous cases that I've seen fragments of previous cases that I've seen instead of concepts. Now, they may articulate the concept. You may ask them when it's competitive, and they give you a definition, they give you a neat thing, they give you a framework, but the way it's represented in their heads is a cluster of cases that they've seen in all the various ways, different various ways that that principle expresses itself in the real world. That's how they're able to operate. Right. The second claim that the theory makes is then that these experts, when you sort of study them, you realize that they all share like a certain worldview, which is that

an ill-structured domain, there is no one cause there's no one lens, there's no one framework, there are multiple valid causes multiple, sorry, multiple plausible causes multiple plausible lenses of looking at it, multiple frameworks. And they are not beholden to anyone. They're able to sort of like, okay, I don't, we may never really know the full set of like causes for these symptoms or like this, this business problem strategy problem, but we have multiple lenses, and they're all equally valid because they all give me information, right? And one of the strong examples of this is that like most of us, we tend to reduce down to a prototype right? So in our heads, we have a prototype of an idea Heart Attack, we imagine a person clutching your chest and falling to the ground. Expert doctors do not have this expert doctors instead have a cluster of prototypes in their heads of what a heart attack might look like. And they know to sort of like, look at each of them. Okay, so why did I do this whole long setup? I did this whole long setup, because the recommendation that these researchers done the is like, how do you use this to teach? Right? And the answer is, when you present a concept, like say scale advantages, right? You give the student an example a case. And it must be a case with full story full context. That means like, an example might be, you know, when Texas Instruments in the 70s, Modi's time was put in there, and the more instruments like oh, you know, like, it's really stupid that we charge for prices. So let's, why don't we charge at a loss for chips, so that there's a huge amount of volume capture a huge amount of market share, and then our volume goes up so that we can climb up the manufacturing learning curve and increase yields. And then we maintain this, we chip production for a very long time to recoup our costs. But we can do that now. Because we are now we now have the majority of the market because of our low initial costs. So this is an example of scale, competitive advantage of scale in semiconductors. But then, then after that, the students are like, Oh, okay, that's a great example. Right? And then you give them another example, that's completely different. So you give an example of Maurice Chang setting up TSMC, which is a very different environment when he left Texas Instruments. He laughed because he couldn't be he couldn't get to the CEO role, possibly because of racism, went and set up TSMC different environment, cuz at that point, it was like two decades later, everybody caught on to the strategy, everybody was using the strategy that was changing method, right? How was he able to compete, so

strong, struggled for many years and eventually realizes that Apple comes to them and says, Okay, we want to make chips and NMRs goes, like holy shit, we are a pure play foundry, we can actually like a ride this wave of mobile to a huge volumes, which then means that we can climb up the learning curve and be much better at manufacturing chips in this arm, you know, like, for this mobile, low power chipset, better than everybody else in the world, we just got to run with that and write that, right, because it's the same principle. It's a very different environment now. And our competitive advantage is that we are like on the position as this pure play foundry where we don't design our own chips, we work with you, and then we build, we try to get as much volume as possible. So we climb up the manufacturing learning curve. And then we've much about the scale advantages than everyone else. Now, different expression, because different time, different competitive set different dynamics, right? They're not TI, right. It's completely funded by the Taiwanese government in initially, right. So you, you give them a concept, and then you give them an example. And you give them a second example, that's very different than the first. And you give them a third example, that's very different from the first two like Netflix, right, Netflix has scale advantages, because it's the largest install base amongst all the streaming providers. And this allows them to have a lower per unit cost to invest in content. So as a result, they ramped up their junk bonds, like they issued huge amount of debt. And this was very risky, and everybody thought they were crazy. But at this point, now, they have scale advantages, because they all that that all that money went into content production. And so now they have this huge just sitting on top of this huge amount of IP. And that they can justify spending on because they are at the pole position, with the largest market Chairman's or the streaming providers, again, a very different example, right. So you give the student as many the authors recommend 10 to 20, different instantiations of the concept with full detail, because in the real world, you cannot separate the detail from like the instantiation of the concept, the student immediately becomes like aware that hey, not this is not that simple, right? If I want to be a business operator, who can use the competitive advantage of scale economies, I now sit with a much better understanding that the way this concept is instantiated, and use is very variable, depending on the on the situation, and the capital available to you and the competitive environment. And then they're able to work like experts in such domains, because now we have like 10 or 20 cases in their heads, they're able to sort of like draw fragments from them. So when they're actually it's time for them to implement a competitive advantage from a scale economy advantage to see the opportunity, they know how it actually looks like, and how it has actually been instantiated in the past. Anyway. So that's incredibly long explanation for like, one way that you can scale this right, which is that you teach students by giving them the concept, and then like telling them go and go and search for the cases. And unlink the concept like put in a collection. And you when you're looking for cases, you want to look for cases that look very different from other cases that you currently know, you focus on the differences, because then you have a much larger set of like case concept instances. With much more variety, which then makes you more effective, because you're able to, like, you know, combine fragments from all these previous cases.

Evan Tan  1:05:09
I mean, what do you articulate? It sounds to me like business schools are Yeah, right. They keep pushing studies.

Cedric Chin  1:05:17
But the thing is they don't, they don't tell you that. So I talked to like, a person who, who he runs an expert network. And he's talked to some of the people in Harvard Business School, on how they write their cases, right. And he said that, like, they don't actually think of cases like that they think of cases as expressions or abstract concepts. And when they write the case, they usually write with an eye to the abstract concept, right? And they don't tell you as a student, like, look, the right approach to reading the case is, you're not there to generalize abstract principles. Because you can't, you're there to see it as instantiations of concepts. But then you don't try to reduce, you shouldn't reduce the principles, you should just hold the case in your head, in all its detail, glory with all the nuance and the mess. With the understanding that in the future, your brain will be able to draw fragments, when it's time to like, implement, you will reason like, oh, this reminds me of that time when so and so this right, which, by the way, is how Charlie Munger tends to reason. So this is a this is one in my sort of essay I sort of wrote, when I read his theory, it is explained to me, one of the longest mysteries that I've had, which is like if first principles thinking is the ultimate arbiter of good thinking, right? The ultimate gold standard, right? Why is it that Charlie Munger seems to reason so much from analogy? He likes to say things like, oh, this reminds me at a time when you know, Coke and Pepsi were in this price war, and then like, Pepsi did this. And then eventually this happened. And you're like, I see the point that you're making, but why are you telling me the story of Coke and Pepsi? Why aren't you just making the stating the concept, the point that you wanted to make, which is that if you have like a substitutable products, it's actually okay. It's actually very effective to go into a price war, but they might not turn out well for you, because it didn't work out well, for Pepsi, right. But it turned out that like, you can't articulate just the abstract concept, because it really depends on the actual example, in which the concept came about. And by telling you the analogy, which manga does, you are able to sort of like, reason more critically about it like this, you really apply this or not, usually, some parts apply, sometimes not. But you have to sort of like, reasonably more critically about it, to be able to use it. articulating the concept, the abstract principle, makes it sound like it's universally applicable, which is not the case in business. As you know, I know, we know.

Evan Tan  1:07:36
Interesting, though, with what you just shared. It sounds to me like there's a trade off going on here. Right? And I'll be curious to get your take on this. The trade off between reading widely and getting the diversity of cases versus trying to be targeted in and identifying what is most important to know, how do you strike that balance? How do you navigate that? Perhaps they're not even mutually exclusive.

Cedric Chin  1:08:11
Yeah, perhaps Perhaps they're not mutually exclusive. Like, the reason I'm interested in a lot of these business things is because like, I see commonplace as this temporary thing where you're along for the ride as I'm trying to improve my business expertise. But at some point, I'm going to actually go and implement the business expertise, like actually go and execute, right? And pretty much the cash flow from the blog is it's meant to like fun, the software or like whatever other business activities that I'm doing. And when you see it in that light, right, like, okay, so this is the way that experts in the field, are able to reason about business moves, then obviously, the right thing to do is to go and get as much cases as possible. Maybe not that many, not not as much as I mean, if you think about your I reading books, just to collect the cases, so you can construct fragments, and no one has come before. It's not very expensive, right? books don't take that long to read. You just have to like not watch Netflix. And I guess, yeah, you're right, in the sense that like, you need to focus on a few things, right to be good at business and, or whatever it is you want to get good at. And if if the thing that you're looking for is not ill structured, then you should just like go and study the concept or like, you know, go and find someone who's good at it and then learn from them. Right. But in general to be effective at your role, you tend to need a couple of even marketing to sort of understand like What a good marketing campaign looks like that looks very different in different companies in different industries. Right? So it's probably worth it to go collect a few samples. And I think most of us know to do this, right? We do this by asking our friends or good practitioners that we meet in conferences for stories about what they've done in their company. And then we go like, Oh, so I know that like Julian from so and so company did this way. And I know that Sara from so and so company did it that way. And then we collect it, and we hold in our head. So when it's time you're like, you know, this sounds like what Julian did, I think we can try can give it a try. Right? So this theory merely argues that experts do this. And the reason they're experts is because they have a much larger case in their heads, sorry, case library in their heads. And therefore the pedagogical recommendation, which is something that I think centrally can do that, you know, any in my in my buckles I wrote, like any note taking app with backlinking capabilities can do is that you can go and find when you quit cases, and rapidly build up this case library in your head, which then means that you can accelerate your expertise because you don't have to be like the expert where you have to wait along and spend many years of experience to collect the cases that then make up your case library in your head.

Cedric's approach to knowledge management

Evan Tan  1:11:15
And that leads me to your next my next question then, would be, you've consumed so much knowledge and information. What is your process? Like? What do you use to document? How do you organize these ideas? What are some of the tools you're using? How do you approach this?

Cedric Chin  1:11:35
Um, so I use one of the, after reading the CFT body of research, right, I realized that I need to create a CFT learning system. So I probably need to go find a note taking maybe obsidian which I already have installed. It's just I don't use it that systematically now I have a reason for like, how to use it and why learning theory so I should probably go and get it done. But the way I do it is actually very slapdash. Right so sorry, every every nonfiction idea book so I split books, I categorize books into three buckets, right. There's narrative, there is sorry, there is nonfiction idea, there is nonfiction narrative. Sorry. I categorize books into three three buckets for non fiction books. So there is narrative again, nonfiction because fiction is also narrative, right? But it's fiction nonfiction narrative, which is like biographies, histories. There's nonfiction idea books, which is like you know, at the atomic habits by James clear. And then there is an A Cody's brunch books, especially they just express one idea. I think I know right, atomic habits. So I don't actually know there if it's been two, or three. But a good example of a bunch book would be I guess, never split the difference. Is, is a good bunch of book predictably irrational by Dan Ariely. It's got a bunch of book where it's just expressing one idea. And then there's three books right, which is like it explains it explains the first principles of an entire domain, like high output management, or principles by Ray Dalio or salt, fat, acid heat, heat for cooking, that sort of distills all of cooking down to just four principles, right? So brunch book, an entry book, and then like narrative, or all, narrative, books, fiction and nonfiction get written on my Kindle. And I don't really highlight when it comes to fiction, because I think stories they stick of your brain a lot easier. And I usually read it to wind down nonfiction idea books I highlight within iBooks. So all nonfiction idea books would be tree or branch. I read it in iBooks, which is automatically synced amongst all my Mac devices. And I highlight an ice and I make annotations. And in the very last annotation of every chapter, which means like, literally, on the very last page of every chapter, I will make a rubbish annotation, literally the last in the last page of every chapter, and that annotation, I will summarize that chapter. And then what allows that, and the summary doesn't have to be you know, it doesn't have to be incorrect, incredibly well written and comprehensive. I could just like, make a bunch of like, notes dash point form of like, what I found notable about that chapter was the main thrust of the argument of the chapter and what was interesting. And the reason why I do that is because later on when I come back to the book, right, I can jump to the end of every chapter and immediately see what I found notable. And I quickly skim through all the annotations and all the highlights. So that sort of like gives me a hit note for every chapter as I hit known as a lawyer, lawyer. When lawyers read cases, they don't want to reread the case. So after they read the case, they tend to write a hit note at the very top that summarizes the case, so that they can refer to the head note and start rereading the entire case, right. And I do that for chapters and I, for some books, I do that for the entire book as well, I like to have a separate document that I write the entire thing. Academy papers, this the same thing that's highlighted in like the in the abstract I will have, I will have a highlight in the abstract, which is usually the first part, the paper that has the overall summary of the paper. So, and for PDFs, which includes papers, I use an app called documents on my Mac OS or my iPad, which I can then export with all the annotations. And so that explains like how I'm able to sort of compress and still be able to refer to huge amount of like, books that I've read, or papers that I've read. They're all in sort of people's own Dropbox, or actually both books and people's in Dropbox. But the books are, if they're in iBooks, they're already synced across every Mac device, that Apple device that I have. And then a lot of my notes are actually rubbish, they're sort of throwaway, I don't really look at them. The if it's valuable enough, these days, if it's valuable enough, it tends to be summarized and synthesized in a blog post format on commonplace. And then that becomes like the way that I go and read my synthesize thoughts. Right. And I think that's it for podcasts. Because I'm happy to see them as well, I tend to use overcast and I clip a certain like the maximum one minute 32nd clips off, like important fragments. And then I sort of remember it. And to be honest, a great deal of this is in my head, right? So I don't actually need a secondary like it's it's both. It's both partly in the annotations, but also partly in my head. And the reason for that is because at any given time, I have like a handful of questions that I'm actively interested in investigating.

And sometimes the evidence for like the answer to it emerges out of nowhere. So like the answer to like, Why does Charlie Munger reason by analogy is something that I've been wondering for like three years now, and only recently sort of resolve. All right. Other questions that I've sort of interested in is like, how for the companies that grew without external capital, how did they actually do it? What are the what are the what are the ways and like, go look for as many stories as possible. Another question that I've been investigating for maybe five, six years is that why are most of the large companies in Asia conglomerates? Right? And they are like multi business, multi industry conglomerates. What happened to in the West, they made them NT conglomerates, right to the point where they had ethical. So anyway, my point is that these questions are like organizing mechanisms for the information that I seek. So I'm not reading with no purpose. Usually I have a I have a specific reason that I'm reading a book specific team that I'm investigating. I'm not sort of like reading for like, oh, look, this is an interesting story about like, sometimes they do, don't get me wrong for fun, right to relax, right? Like reading Elon Musk's biography was like that, for example. I was not very interested in Elon Musk, but like, fun biography and very cool story. But generally, when I read something, it tends to be for a distinct purpose. blog posts are the same thing. Yeah. Is that

Evan Tan  1:18:39
a very conscious process for you, you have this question in mind. And then you read that you document what what actually is relevant to this question. Do it that way, or

Cedric Chin  1:18:49
it's more like everything I read. I tend to find Fineman does this as well. I mean, Simon talks about how he has like, the 10 most important questions in his head, right. Every time he sees a new piece of information, he asked himself, is this relevant to one of my 10 questions? 10 open questions in physics. I don't I mean, it doesn't erase him. I, I've talked to a number of people who are quite who who write with the same level of rigor than I do. And they seem to have a very similar sort of approach as well. Researchers as well seem to think like this as well. intellectually curious people, if you're, I think if you're rigorous enough, about the way you investigate things, you tend to eventually stumble into this because they like certain questions that have a grip on you, right? And you're like, how, right? So? Yeah, so anyway, when I look at a new piece of information, I like my brain automatically goes like this is relevant. Usually, it's not even a conscious process is very immediate, like, Ah, this is why among the reasons why not yeah, yeah, exactly.

Evan Tan  1:19:57
And is retrieving the notions ever a challenge for you? Or well, because

Cedric Chin  1:20:03
you, if you have a burning question, it's always like sort of at the back of your head, right? Like, you could put me in a beach. And eventually, if I'm staring out there, and I'm not in present in the moment, I, my brain will take one of these questions. Like, why or why?

Evan Tan  1:20:20
I'm just, I'm just trying to imagine your process right now with all these different notations, these summaries, floating all around, and Dropbox and all these places, and then the organizing questions in your mind that you're trying to then connect all these dots together?

Cedric Chin  1:20:39
So most of it happens in my head, but I'm fully aware that you synthesize it when you write it right on the books. Yeah, well, not. So I synthesize it, a lot of the synthesis actually happens. Ideally, it happens before I write. And the way it happens is through conversation. So the reason why I could distill CFD down to like the four big ideas, is because I had like 20 different conversations with the cause of January, where I was trying to try to articulate in the best possible form, right, the best, most smooth, most easily digestible form, because the idea the research literature was vast, and I think I read like, three or four different papers and like, two to three different like precursor works. And, and each of these papers were incredibly dense. And they called out to like various other like bodies of literature, which I then skim. I didn't know how to articulate CFT, right? And talking to people, like clarify my tops, because then you will, I will see where people are confused, I would not be very happy when like, I can tell that the way that I'm explaining things are not very clear. And so like, what I'm trying to say is that, what we'll do things First of all, like you synthesize by explaining, and then when you write this, like sort of the ultimate arbiter, or the ultimate standard for like, clarity, or because you cannot lie to yourself, you can, there are tricks, but as a writer, you know, if you're trying to like use tricks to cover up holes in your thinking, and don't get me wrong, I do do that sometimes in common Kog. Mostly, I don't say I'm not happy if the question is important enough. The other thing to note is that I do a lot of this in my head. But I know from talking again, because I talked to some of these people who do something very similar. They tell me that wants to you have kids, right? Once you get married, you have kids, this is impossible. And so you need a system to do this for you, more externalize thinking and less in your head. And I believe that I truly do believe that kids take up a huge amount of your time. And therefore your cognition. And I think whatever I do is not tenable. Long term, right? Unless I don't have kids. Yeah.

Evan Tan  1:22:51
So if you have kids, then you definitely need to start externalizing all that stuff.

Cedric Chin  1:22:57
Right. Yeah, yeah. So So So I mean, if you look at all these people who I think are similar, and then they have kids, and then it takes something, it's usually quite wise to like, heed the warning. Gotcha. Yeah. All

The possibility of collective mental models

Evan Tan  1:23:11
right. Well, I think it's been fantastic. Everything you've shared all the ideas, the ground we've covered, I'd like to sort of wrap it up with maybe a question to you. And for you to sort of approach however you will. One of the ideas that I've been exploring together with my team that's really around, how can we try to map out some form of a collective mental model, some sort of knowledge graph that's overlaid onto a social network? And what what what would your thoughts be around this notion of, can we actually try to, you know, take externalize our own map of concepts of the world and integrate it with other people? And how what comes to mind when you think about that?

Cedric Chin  1:24:11
So you know, in know, this dream, is a huge one in the tools-for-thought space. I also know that there are some precursor works. So like Wikipedia is like the ultimate example. Right? But that's because it's a it's it's taking a form that is well known the encyclopedia and it's making it hyper linkable and user generated. And I think Wikipedia has done a good, a lot of good for the world, right? So then the question is like, how do you do this better? I think doing a case soap. CFD was really notable to me because it was the first time somebody had articulated like how a case library could actually help you know, Of course, you must tell the student and the people reading the case library that like, hey, the way to use this is to understand that you're collecting fragments, don't try to reduce this to abstract principles. And then there's a few other like things which you can go read my essay sort of see, like the tips on how to make such a graph of cases more useful. And he they purposely say, backlinked, you know, note taking graph. And, and I think that's the biggest way that you can do it. I am very skeptical of this idea that you can create a knowledge graph of mental models have collective mental models. And the reason for that is because mental models are not explicable. So let me be more precise, the military has spent a huge amount of money on the with the researchers in naturalistic decision-making a specialist in exactly this, because the military wants to be able to extract this tacit models expertise from the heads of his soldiers, so that he can then train newer soldiers without, without actually sending them to the battlefield and like learning it through blood, right and death, right. And so, the best, and you would think that they are very motivated, they have a huge amount of money, there are millions at this, there are some really very smart researchers who have worked on this problem. They have done been doing this for 30 years. So you would think that they are further along amongst everybody, right. And one of the things that I'm trying to do, and that I've been telling them, my friends in the to forethought, space is please pay attention to this, right? It doesn't matter. I don't I don't, whatever you come up with, I guarantee you the military probably has thought of it and has tried to put into practice, because they care a lot more than you do. And they have a lot more money than you do. Right. And what they found is that the best that you can do is you extract the mental models of expertise from the hits of these people, you put it into a the the ability, the technique to do that is Cognitive Task Analysis. By the way, at CTA, you put that into like a there's a formula, a number of formats of like how to collect this research, right? I think it's like, it's basically qualitative research. And a lot of techniques are borrowed from like standard qualitative research, organization methods. But that is not the result of the CTA is not the thing that gets used by the military. Because that's only useful to the organizational psychologists who did extraction, how they implement and make sure that the tacit knowledge is communicated and lives inside the organization is that then turn it into training simulations? So the Marines at this point, have a library of huge number of decoy tactical decision games are I think there's another term for it, but like TDG is one definitely one of them tactical decision games, where they are scenarios extracted from the heads of actual Marines, or synthetic scenarios, designed with the help of expert Marines who have come back from battle from from from from actual deployments. And then those tactical decision games then use to train and help. Newer, and usually the way it's done is I give you the technical decision, decision point want, you know, I give you a scenario? What do you do? Or like, what is the most important area to take control off? Right, the bridge healed? And then you answer, and then we will reveal to you the expert answer. So that's like one very scalable way.

A more loose, less scalable, but equally rich way is to have an actual expert practitioner in the room where they're doing this. And then they give everybody gives the answers. And then the expert Petitioner articulates their answer. Or sometimes the Marines can have like two expert practitioners, I mean, experience commanders in the room, and then they might give different answers in a debate, debate and analysis watching, get a lot of information on it. Now that that is the gold standard for where the state of the art is right now. Right? And the the reason is very obvious why right? If you extract the tacit mental model into something, explicated and written down, stored in some database, the problem is Nope. And they've tested this, they've tried this, nobody goes and retrieves and reads it. They don't, it's too busy. It's not built into the order. What kind of marine will go and read up the results of a CTA, right? They're too busy. No, you need to turn into something live and, and they have successfully used this not only in Marines with in a military context, but also in industry context. So like, one of the stories in accelerated expertise is a story of an electric company, where they had expert electricians who retired and then when a car crash into a very quite a poor electric pole in New York. And there was like five generations of systems on that one pole that broke no amount of the current electricians or technicians could engineers collect, figure out how to get it to work. So they had to call an old person who already retired for like maybe like five years right to come back and say, Okay, I because I was here for most of it for the five decades where they slowly add a new immune systems, here's how you fix it, right. And then the electric company was incredibly scared. And then they called in some of these MDM researchers to do the tacit knowledge extraction, and in how they made sure that it lived on in the organization so that these people can retire and they can pass on without, without worrying that the knowledge dies with them, right? Is they turn it into simulations, they turn into training programs that all the members of the the actual technicians of this, this particular organization then had to run through as part of their job requirements. And that's how they kept make sure that the knowledge continues to live. And because very easy to update this, because then you just, you know, if things change, you just do an extraction, you create a new set of simulations, you add that to the library of simulations that all trainees must go through, and then the knowledge continues to live. Right. So this extraction thing has been done in the I think it was the 80s or 90s, the idea of like, extracting it into a library, a knowledge base that everybody can search. And then the reason it failed is because nobody searches for anything in the knowledge base. So this is the current state of the art that the military knows the military uses.

Evan Tan  1:31:11
I think when you're out there, it's so interesting, right? That it's one thing to extract the knowledge, unless you're able to package it make an expert. It's it's not going to stick? Yes. Yes. Yes, that's really the trick.

Cedric Chin  1:31:24
Yes, that's the real trick is making it live. So you can extract the tacit metamodel. But then the creation, the passing on of that mental model to other people, right, requires some sort of like, it's not written, it's usually some sort of simulation or some sort of activity. I think the craziest idea, I know, we're hitting up on time, the craziest example this story is, and I talked to one of the researchers who worked on this is that is the story of the the ID detection, right? Where they had, they had some soldiers who could recognize IEDs, or roadside bombs in Iraq and Afghanistan, right. And the military wanted to know how this, these soldiers were able to recognize when the possible ID was on the road. And, and they managed to successfully extract the mental model or the passive mental model expertise from the hits, the reason why they had a bad feeling like don't go down that road, it's probably a bomb. So they successfully extracted it. But then the way they thought this to the new recruits were going to be sent was they didn't give them a lecture. They didn't write it in like a bunch of case studies. No, what they did was they designed a computer game, where the novices were had to play as the terrorists, or we had to pay as the insurgents to set up the IDs, because that captured the true new ones of what the actual soldiers were doing, which is that the expert soldiers who were able to have a bad feeling were putting themselves in insurgents shoes. So they thought this and communicated this mental model by forcing all the people who were going to be deployed in Afghanistan or Iraq to play this computer game, like a couple of rounds of it, just to get the metamodel of how the insurgent thought and all the constraints they operated under when they were in placing an ID. Right. So so, like, yes, you can extract but like, how do you communicate? That's the problem. How do you construct the metamodel in the heads of the novices?

Evan Tan  1:33:19
Thanks so much, and I think we could keep going on this. Yeah, we can be exploring, but uh, once again, thank you so much for coming and being my first ever guests. And yeah, I'm looking forward to doing more with you. Maybe another follow up session, and we can take a few more of these ideas.

Cedric Chin  1:33:38
Alright, it's my pleasure. Thank you so much.

Transcribed by https://otter.ai