Because the focus of this learn better idea is really, you need to learn how to learn. And once you have that tool in your tool belt, you're pretty much unstoppable, right? Then you can just apply that to whatever content you need to learn. So, literally, my goal as a teacher is to expose students to as many possible ideas as possible, because the earlier that you can learn things, the more that you're then able to learn in the future. Hello, and welcome to the PY Bytes podcast, where we talk about Python career and mindset. We're your hosts. I'm Julian Sequeira. And I am Bob Baldebos. If you're looking to improve your python, your career, and learn the mindset for success, this is the podcast for you. Let's get started. Welcome back, everybody, to the Pibytes podcast. This is Bob Baldebos, and I'm not with Julian here. He's doing other stuff. But I have a very special guest. Russell Helmsteader is back on the show. Russell, welcome. Hey, Bob, how are you? It's good to see you again. Awesome as always. For anybody listening, we had Russell back on episode 67. Russell is a teacher at Data, or let me say Ventura Unified school district. Yeah, yeah, yeah. Data is de Anza Academy of Technology and the Arts. You should definitely listen back to episode 67, where he and Alex Wolf go very deep into what they do. It's super inspiring, super motivational. But anyway, welcome back. Today we're going to talk about learn better, a book we have been reading while you finished it. I'm still reading it, and it's a game changer, I think you shared some very detailed notes with me, and there's a lot in the book we can pick up and become better learners and be more effective. There's also, surprisingly, some stuff we want to link back to Pybyte. But before we get on into all that good stuff, do you want to share win something that's going on at work? Let me know. Yeah, absolutely. I'll actually start with the win of our academy. So when we spoke on that podcast episode last time, it. It actually did not exist yet. We were looking forward to doing it this school year, and we're. We're doing the thing. It's awesome. This is. This is year one. We have our first group of 6th graders in. We're doing the computer science elective. So we're learning Python in 6th grade. They. They have my colleague Alex Wolf for math and science, which is fully project based. CNC routers, laser engravers, 3d printers, all that good stuff in a math and science class as a 6th grader. And, yeah, it's just. It's a really exciting time. And the energy this school year feels really different. It's a really good energy compared to what it was especially two years ago with distance learning and COVID. So I would say that's a really, really big win for us this year. Awesome. Super happy to hear. And super exciting. It's all happening, right? Yeah. Futures here. Yeah, last time that was all not there, and it's all happening now, so really cool. Yeah, it was just an idea at that point. And now it actually exists. And again, for what we're talking about, go back to episode 67. There's a lot more context to that. All right, if you're ready, shall we dive into the book? Let's do it. So I had stumbled across this book. It's called Learn better by Ulrich Boser. I hope I'm pronouncing his last name correct, but I stumbled across this book when I was in my master's program for curriculum and education, and it actually stemmed from an NPR article that was talking about the myths of learning. And so the article is really cool. Maybe we should link to it as well. You can just take a look. It's a really short quiz almost, if you will, of just like, hey, like, what's the best way to review material? Do our highlighters effective? Turns out highlighters aren't that effective. Right. And so we. We kind of read this article for the. Our master's program, and then at the. At the end, it links to this book. And I was like, oh, that's really cool. I should read that. And then fast forward two years later, I finally read the book. So once I actually started reading it, I couldn't get enough. It's definitely already changed the way that I teach. And I've shared parts of the book with my students already because the focus of this learn better idea is really, you need to learn how to learn. And once you have that tool in your tool belt, you're pretty much unstoppable. Right. Then you can just apply that to whatever content you need to learn. Yeah, it's about the meta skills. And, yeah, I'm 50% in, so I still have a. Was highly addicting. And I'm making a lot of highlights, which kind of shows that I'm using the wrong way, but I will go back to them and make notes. But, yeah, so the structure, it's all about the six facets of learning, right, which are value target developers, extend relate, rethink. So, yeah, I think we could go in that order, unless a high level summary. But I think if we summarize those six, we should hit the major components, right? Yeah, let's do it. Before we get started with the individual six facets, you have value, target, develop, extend, relate, rethink. There's an idea that I want to touch on, which, in my opinion, is kind of the glue that holds all of these things together, and that's the idea of the knowledge effect. So, according to the book, Ulrich says that what you know now is often the biggest predictor of what you are able to learn. And when I read that, that just hit like a ton of bricks. It's like, oh, man. So, literally, my goal as a teacher is to expose students to as many possible ideas as possible, because the earlier that you can learn things, the more that you're then able to learn in the future. And so this idea of the knowledge effect is that once you have these facts and ideas stored in long term memory, anytime that you're trying to learn something new, you have context for that new knowledge to seat, and that just becomes so powerful because you have a deeper and a wider pool from which to think about relationships and new ideas and new concepts, and that will just compound your learning even more. Wow. So, yeah, you're creating this reference, right? And every new thing you learn, it's easier to make links, and therefore, you will learn faster. Right. Not only faster, but also, I would say, deeper. And so it's not all about speed. I think that there's a lot of value in slowing down and thinking deeply about things, but it's not like our brain is a computer file system. It's much more of a web where you have these connections, and if you can strengthen those connections between ideas and concepts, that's where you get into that really deep, deep understanding. What's really difficult is that before you can make connections and relate things, you have to know them, right? And so these ideas of facts and fluencies matter. And because otherwise, you can never reach flow in your thought process, in your problem solving skills if you don't have that foundational skill. But then if all you have are facts and you just did rote memorization, you still aren't able to connect those things, and you really haven't reached that deep level of understanding either. Interesting. So it's all about finding that balance between big ideas and memorization, because I think both are important, because, sure, we live in the time of Google, and you can just look up the answer to pretty much whatever you need. But if you're constantly looking up every single part of python syntax, how are you ever going to be able to code? Yeah. Right. You're not going to be able to construct complete programs, which has a lot more to it. Exactly, exactly. So with that kind of glue, the knowledge effect, holding all of this together, let's jump into value. Let's jump into each of them. So value. So value is essentially closely tied to motivation. And Dan Meyer is a math teacher here in the US. He has this quote that I absolutely love. He says, I am a high school math teacher, and that means that I sell a product to a market that doesn't want it, but is required by law to buy it. Right. And if you are a math teacher, at least in the US, this just, this stings because it's so true. Right? Like, it's really difficult some days to try to show students the value in solving quadratic equations or, you know, like, why do I have to memorize this thing? When am I ever going to use this? And so it turns out that if you don't value the content that you're trying to learn, you're really not going to learn it. Right? Sure, you can be exposed. Sure, maybe you become familiar with it, but really you're going to have to struggle with it and grapple to reach that nuanced understanding. And if you don't find value in that, it's going to be really, really difficult, if not impossible to learn. Yeah, I think we all can relate back to when we were studying materials. We're not engaging. Boring. Why do we need this? Right? And we like that, motivation. Um, and, and then the book also say, like, simply telling the learner that the information is important, it's not enough. Actually can, can backfire you. Right? Yeah. And that actually ties in closely to Dan Pink's work on motivation, where you're, you're losing your sense of autonomy if you just have somebody screaming at you, this is important. You need to do this. You no longer have the choice, and that can absolutely backfire. So then that becomes a predicament, especially when you're trying to educate a wide audience to say, hey, this stuff is important, but now let me actually show you how it can relate to your life. And that doesn't necessarily mean that it needs to be this quote, unquote, real world application. It could be very abstract. But if it's interesting, and if you can peak people's curiosity, that's oftentimes enough. Awesome. Yeah. Cool. Let's move on to target. Awesome. So targeting is really about setting goals for yourself, creating plans, and then focusing on mastery of very specific concepts and ideas. So if you say, hey, my goal is to learn Python, that probably isn't very helpful. Python is this big, wide pool that my gut instinct tells me. There's not a single person on the face of this planet who knows everything about Python, probably not even Guido himself, just because there's so much. Instead of saying, my goal is to learn Python, maybe if you're a beginner, your goal is to learn the syntax of Python. And all of a sudden that becomes something that is measurable, it becomes something that can help you focus your efforts. And then once you've kind of mastered that, and we can talk about that more in the developing facet as well. Next. But once you've kind of achieved that mastery, now you can change your goal. Okay, now that I know the basic syntax of Python, maybe now I can study more advanced methods for strings, for lists, for sequences. You start targeting, you make goals that are measurable, and then you move on to the next one after you've mastered, for example, on our platform, if you're a beginner python, you go through the newbie bytes. You have 25 bytes explained in the fundamentals. Thanks again for doing the videos for those. Absolutely. And then you have the learning path as well. Right? Like you want to learn algorithms or you want to learn web scraping or data analytics, right? You have a very concrete path. These are the 20 exercises you need to do, and it's measurable. So. Yeah, like having specific goals and breaking down goals into digestible steps. Right. Is very important. Otherwise it's what doesn't get measured, does it get managed? Right, as Peter Drucker said. And, yeah, absolutely. I totally agree with that. And then in addition to that, when you actually think about, you know, this goes back to the programmer's brain, what you and AJ spoke about in that, in that podcast episode two weeks ago. Yeah, yeah. Where you're trying to get information into your short term memory. Right? And that's all this new stuff that's coming in. And the only way that you can transfer from short term memory, the long term memory, is if the information that you're digesting is small enough to be stored in short term memory. And so going back to that python example, if you're like, I need to learn all of python, and you're drinking through the fire hose where you just have all of this stuff and refactoring and formatting with black and syntax and all these other things all at once. The end result is going to be that you learn nothing. Sweet developer. Perfect. Let's jump ahead. So this is after you've made your targeting goal, you actually need to practice. And so I don't think the boser ever uses this phrase exactly in the book, but when I read this, this just screamed of deliberate practice to me. And if you go look at the literature on deliberate practice, there's essentially four main components that you need. So, number one, you need to have a task with well defined goals. Oh, the targeting that we just spoke about. Number two, you have to be motivated to improve. Oh, that's the value component that we just finished talking about. And then you have to be provided with feedback, and you have to be provided with ample opportunities for repetition and gradual refinement. And if you meet those four criteria, that is deliberate practice. Right. And, you know, there's. There's a lot of. Maybe controversy is a little too strong of a word, but there's some pushback against this idea of 10,000 hours makes you an expert, or, you know, practice makes perfect. Like, those kind of ideas are like, well, wait a second. If you practice for 10,000 hours but you practice incorrectly, are you actually learning things? Probably not. That's a good point. So the two big components to me in this develop thing are getting feedback and then having that opportunity for not only for repetition, but also for refining how you're doing that repetition. And to me, like that is pivots. That is the coding platform where you have instantaneous feedback with the tests. You know, if your code works or if it doesn't, you get that instantaneous feedback. Then you go into the slack channel and you post a question and you're like, hey, here's kind of what I did. Here's how I solved it. What do people who have either maybe more skills than I do or maybe even just a different perspective than I do, how would you approach this and getting that, that repetitive. Look at it. Getting that gradual refinement and then going back and refactoring. Like, that's it. That is that developed stage where you're constantly working towards mastery. Happy to hear that. Yeah. And a book on deliberate practice. I think you linked the author, Ericsson. Isn't that from the peak book? So the Ericsson link that I have is actually a research article. It's not even a book. Okay, so it's his actual research article where he created a situations with those four criteria in place, and then he tracked progress of people and how they were learning. Right. And so there's a quote from that that I put in there, which is the best training situations focus on activities of short duration with opportunities for immediate feedback, reflection and correction. Like short focus duration, like bytes of code perhaps. Right? Like it just, I think it, I think it works really well and I think that's why, that's what drew me into the platform in the first place. Right? Like I was a, I was a teacher looking for a way to target and provide opportunities for my students to code a specific skill. And that's what pie bytes lets my students do. I love the feedback that they are quite addicted to it as well. Yeah, yeah, they, and honestly, it surprised me how much they liked it because it's not, it's not super visual, right. They're not building video games and it can be a little intimidating. I don't know if you, if our listeners have ever seen like the pytest output, right? You click that and all of a sudden you just have this wall of red text. And for an eleven year old to see that, a lot of times they're like, oh, oh gosh. Oh gosh. They just like close the computer, push it away. Yeah, it's, it's absolutely just shocking. So I think it's, I think it's really exciting for me how much they actually get into it, especially once those things turn green. Oh man, that is just, that's the best feeling. Yeah. We're super excited hearing that, how schools are having success with our platform. Enormously gratifying for us as well. All right, let's move on to extent. Awesome. So now that you've targeted what you're looking for, you've set those goals you've been developing, you've worked through your deliberate practice, now it's your turn to extend your knowledge. So you extend mastery by synthesizing ideas, right? So it's not enough just to read information and highlight information and then go back and reread it really in the best possible world, you are working on elaboration. You're taking the things that you know and you're putting it into your own words. And Boser offers this idea of the three sentence essay. Can you read through a paragraph or two paragraphs of text, something that's containing an idea or a concept? Can you put that concept into a three sentence essay that explains it at a deep level? And if you can't, you probably haven't learned it yet. So then at that point you maybe you do need to go back and reread. And maybe at that point you do need to highlight. But if you stop there, that's not enough. Right? Like, take it further and synthesize those ideas, really, really own it and make it your own. And then the next idea in this extension level that as a math person, I just love, it's this idea of squeezing and stretching ideas. And so when you squeeze an idea, you zoom into it and you focus on the minute details. Right. And by focusing on those tiny little details, you start developing nuance and you start developing a deep understanding. And then on the other hand, you need to zoom out and stretch this idea into an abstraction where now you're not worried about details at all, you're only focused on, okay, how can I generalize this idea? How can I relate this idea through analogy to something else that I already know? And so when you do that, you lose the details and the nuance, but you gain this relationship or this relational understanding of where this fits into the bigger picture. And I'm just like, oh, man, that's a beautiful idea. And that's kind of at the heart of mathematics. We would look at very specific cases, and then we zoom out and make these generalizations of mathematical proof. And so if I can take that idea into everything else that I'm learning, like, that feels right to me, and that feels good. Interesting. And you implemented this in your. Yeah, your system, right. Of note taking use Zettelkasten. So that's a note taking system. Yeah, focus on linking stuff together and, yeah, we'll talk about that in the next facet because that's the relation part two. And again, it's not like each of these facets are completely independent and never interact with anything else. That's not the case. There is going to be overlap between extension and relation and targeting and developing. So there's always that interplay between them. But I still think it's useful to squeeze these ideas down into the six different facets and then zooming back out, stretching out the idea and saying, oh, this is actually the overall umbrella of learning how to learn. The last thing that I do want to touch on under the extent is this idea of pattern expertise. There's some research that's been done that talks about the difference between novices and experts is that experts have the ability to identify patterns. They have a very strong knowledge effect. So they have all of these facts and ideas already stored in long term memory that a novice doesn't have. But then what separates the novice and the expert from, like, the truly expert is then the ability to connect those patterns right, so you're learning something new, you're going through and you're saying, okay, so I know that I can make a string, right? And then the extension of that string is now learning how to use string methods. So you've moved beyond the basic syntax and you've taken it one step up to, oh, I can use methods for uppercasing or lowercasing or you know, stripping off leading spaces, for example. And if you're an expert in, in computer programming, you can then say, oh, I totally already know what that is because I know what list methods are. So I know that I can add things to a list with append. Boom. So now you have that pattern of the method notation that you can use to extend your knowledge of strings in python. And so that's what separates the novice from the expert, right? Because as you already know the list methods, it's easier to reason about the string methods. It's the same principle, but on another type. Right, for, of example. Exactly, exactly. And again, like it's, we're starting to get into this like relational territory, which is the next facet. But that's what the perfect, like, that's what the extend is, right? You're squeezing and stretching ideas, you're writing these three sentence essays and then literally the very next idea is you take that extension and now you relate it. So relating is making connections between new knowledge and knowledge that's already stored in long term memory. So the way that you really work on this relation is you develop the pattern expertise that we just spoke about and then you identify deep and shallow features. So deep features versus shallow features is kind of, it's built into that idea that we just spoke about, about methods. What is the deep feature? The deep feature of methods is that I can use methods to, I want to be careful because I wanted to say the word mutate, but strings are immutable, right? So when you're doing a string. Oh, perfect. So I can use, I can use methods to update either my strings or my list or whatever, right? Methods are functions that work on some type of type. That is the deep feature. Like that's the overall idea of what a method is. And then what are the shallow features of that? Oh well, in string methods they don't actually change, they don't mutate the original string. You get a new string that has been uppercased or lowercased where if you have a method like the append method in lists, you are now literally mutating and adding to your appending to the already existing list. And so if you can identify what are the deep features of something that can generalize across different domains versus the shallow features of the very specific problem, you're going to be able to learn new things and understand new things so much better. Right? And we do the same thing in mathematics where we offer. Sometimes it becomes even a trope because you just have this word problems where you're working on these word problems that are designed to get you to work on your addition. And so if Johnny has three apples and Celia has five apples, how many apples do they have together? What are the shallow features of that problem? Well, it's the apples, it's the kids names, Johnny and Sally. What is the deep features of that problem? Oh, like the heart of that problem is addition. And if you can identify those deep features, then you can start solving any kind of problems whose deep features are in that domain that you already know. Fascinating. And do you see a use case on the platform for that? So on the relate part, I would actually argue in full transparency that this is probably the weakest facet of the learning is the relational part. I did go back and look, and you can add notes in there. So after you solve your bite, you can absolutely write down like, hey, what did you learn from this? And I think that that alone is beautiful. That's the extension piece. If we could then find a way to connect notes between each others so that you could say, hey, I'm working, I'm going to make up the numbers right now. Don't quote me on this, but I was working on byte number five, and that was string methods. And then if I go back to byte number three, that was string syntax. If there's a way where the user can then type in there, oh, hey, this is just like what I learned in byte whatever, and they just have a single link that takes them back, then they can see that, and that's how you can start building that relational, that relational context. And I think that would just, boom, take it to the next level. I think that'd be really cool. Is that all custom for PI bytes? Yes, yes. I mean, linking bytes would be easy, but I'm not sure, like we have the tags on the bytes, but that's kind of imperfect. Well, I think it could literally just be a markdown link, right? Maybe, maybe this doesn't need to go into the podcast, but, or maybe it's interesting to people, I don't know, but I think that that would be really cool. We'll leave it in, maybe somebody reaches out tomorrow and like, can I contribute to that. Yeah, like, hey, here's the pr that you guys were talking about before linking. I just built this API fast. API. Perfect. Yeah, I don't, I don't think it has to be crazy. I think it could literally just be a link to the URL where you say, hey, this is what this byte was about. Here's what I learned. This is the same deep feature as this. And if users of the platform can build that into their practice, that would be amazing. And that's why we do use the Zettelkassen. And lastly, we have rethink. So the last facet of rethinking is this idea that familiarity does not imply understanding. And that is huge, especially in a classroom setting for me. Right? I have. It happens all the time where I'll present an idea or a lesson, and kids will say something along the lines of, man, we've already done this. I'm like, oh, okay, cool. Like, would you mind just taking this test on it real quick? Like, it won't be for your grade or anything like that? Just take this quiz, and then they take the quiz and they fail. Right? So, like, yeah, they've, they've seen this before, but have they actually taken the time to grapple with it, to develop that understanding and then. And then go from there? And so the. The bane of this rethinking idea is the. The idea of over confidence. And. And again, just like, I feel like my theme of the day is going to be balance, because with this overconfidence, if you. If you don't have confidence at all, you're never even going to start to learn, right? You're just gonna be like, nope, this is gonna be impossible. I'm not gonna be able to do it, or you're gonna think you're not good enough. Right? Like, we've talked about that before. Like, just ship it. Sometimes it's just good enough, right? So you have to have the minimum amount of confidence to know that you're capable of this. But then if you have overconfidence, you slack off, right? You stop studying, because, like, no, I've seen that before. I totally know it. And I actually had a moment of that for this podcast when you reached out to the idea, and you're like, oh, hey, you've read the book already. Can you come on the podcast? Yeah, absolutely. I could totally do that in my sleep. And then I actually went back and read my notes, and I was like, oh, I had forgotten about that part. Yeah, we need to talk about that. Oh, I'd forgotten about that part. So if I had just come into this podcast cold, without rethinking, oh, wait, am I sure that I understand what the knowledge effect is and how that connects? Am I sure that I understand the difference between extension and relation, but also that they're connected? And so taking that moment to go back and reflect helps you develop that deep understanding and actually get rid of old mental models that don't actually apply. And so I don't know if you're familiar with the Laffer curve or not, but it's this idea, especially in economics, that the more that you tax, the more revenue that you have. But the more that you tax, the less people want to work, because all of their work just now goes into paying taxes. And so the old idea is that it was just a linear line where more tax is bad. And it turns out that probably not. It's probably more like a bell curve, where there's an optimal point between taxation and revenue and people still wanting to work because their work is valued. And so if you apply that idea to this idea of confidence, what does that curve look like? If you have no confidence, then you don't even get started. But then if you're overconfident, you've now gone past the peak, you're back on the downward slope, and now you're slacking off again where you don't actually have the understanding that you believe. So how do you find that kind of local maxima? How do you find that optimal point of confidence to be critical of your own work and still believe in yourself that you can do it? Which I think is really cool. Yeah, I think when we learn new topics, there's this phase of, like, total impostor syndrome. Don't even want to start, and then you gain some skills, and then you actually might become overconfident until you hit a new level or you get a code review and like, actually, totally. I don't know that much about it. So you go grinding again, and so it's kind of an up and down, right? Absolutely. Levels. Yeah. And if we, if we just wanted, the last idea here is to tie this back into the platform. To me, this, this is just refactoring in a nutshell. Go back to the same code base that you wrote that you think you know and actually go through this. I was just working on this project for this salary comparison idea, and I started doing the refactoring, and I realized that I was passing in these arguments to all of my functions that didn't actually need to be there. So I was like, oh, like I could just construct this within the function and I don't have to have all of these extra arguments littered all around. And if I had never taken that moment to rethink it because I thought that I had already solved it the best way that I could, now all of a sudden, oh, it's much cleaner, it's much easier to maintain. For me, there's a lot of power in this idea of rethinking. And then in the context of coding, refactoring. Yeah. You also shared a note about solving the same exercise, but adding constraints. Oh, that idea is just beautiful. It's like if you play a musical instrument and you're just developing a riff where it's the same melodic line, but you just add in a little 16th note, or you just add in a little something new, or you say, okay, I'm going to do this same exercise, but I'm going to do it without any string methods whatsoever. Anything that I'm going to construct, I have to do it manually. You write a little function that does what the string method does, you're like, well, why would you do that? Because that already exists, so you're just wasting your time. Well, I'm probably not going to do this in production, but for my own learning and my own understanding, that is unbelievably valuable to actually go through and rethink that process and like, hey, can I actually create this on my own? And think about what that'll do for your understanding overall, forcing yourself outside of your comfort zone and, yeah, kind of challenging yourself to. To make more connections. And I would argue that it's not even forcing yourself outside your comfort zone, it's diving into what you believe is your comfort zone and identifying points of weakness within that. Right. So it's not even about reaching out. That idea is about really just drilling down into, do I know this? At every single possible level of nuance. Awesome. Well, I think this wraps up a very dense episode. And thanks for sharing all this. I mean, how well you have distilled it and structured is a testament how you made the connections yourself. And thanks for going back, doing that second round. Amazing. Thanks so much. Thank you for having me, Bob. It's always good to talk to you. Yeah, we might as well, we might have a follow up out of this one because there were some things we could dive deeper into. I think we should spend some time about zelikasm. I think we should talk about the paradox of the knowledge effect, but we kind of did a little bit with that rethinking, but also, I know you got to go, and I got to go, but maybe we can just link to the notes for people as well. Oh, that will be on the GitHub repo, and I don't think there's anything in there that's proprietary for you guys. I think the only stuff that's in there is intro bytes. So link the notes. I'll send you the article. We can link the article for that little learning quiz and call it a day, man. Appreciate it. Yeah, people will get a lot of value out of that. Yeah. But you're always welcome to come back. I mean, as you know, it's a podcast about Python developer and mindset. But the meta skill of learning is so important as a developer, but in any area. So we definitely want more of that on the podcast as well. Yeah, let's do it. So thanks again, and, yeah, everybody, go read learn better by Ulrich Boser. Right? Or Boser. Yeah. Seems a german name. Yeah, yeah, totally, totally german name. And maybe that's why I like it so much. Helmstadter and Boser and settle casten. Yeah, perfect. Full circle. Full circle. All right, man, have a great day. You too, Ben. We'll see you. We'll see you soon. We hope you enjoyed this episode. To hear more from us, go to Pibyte friends. That is Pybit es friends, and receive a free gift just for being a friend of the show and to join our thriving slack community of python programmers, go to Pibytes community. That's Pibit es forward slash community. We hope to see. See you there and catch you in the next episode.