In today's episode, I want to talk about some of the ways that our brains automatically parse the world and how that affects our goal setting and how we track progress in those goals, how we think about projects. Before I get started, I have a little bit of a cold, coming out of a cold, so my voice sounds a little different. That's probably why. So part of the way that our brains work, part of the kind of default mode that your brain operates in lends you or leads you to think in, and we're going to cover two specific ways that we think. The first is linear thinking, right? Linear thinking. And the second is infinite or essentially non-discrete infinite thinking. And so we're going to break these down and kind of the implications of each of them and ways that it shows up in your planning. The reality is in this episode, we're not going to give you major details. We're going to give you major tips or tricks for how to overcome these things. Awareness is really the first step in understanding and trying to build a strategy to overcome them. And recognizing that this is true should help you at least assert that in your planning process, maybe add it to your checklist, for example, add it to whatever tools you use to kind of square up with reality. Right. Or try to, you know, fight some of the biases that you may fall prey to. So the first is linear thinking. And we talked about this quite a bit on the show before, but it deserves more attention on a very regular basis. This is showing up, definitely showing up in our jobs today. One statistic related to the progress of AI agents underscores the importance of our default way of, of thinking. Linear thinking also has kind of a sub part to it. I would call it static thinking. So essentially the way that we imagine the world and this is, I'm going to kind of combine all three of these things, right? We have static or sorry, linear, static and nondiscrete or endless infinite thinking. So, so linear and static mean or linear. In a way that all of the, when we encounter something, we imagine that whatever that thing is doing, it will continue doing that at approximately the same rate, right? And rate, you can kind of substitute for at approximately the same behavior, right? And so, so a good example of this is we may take a snapshot of the people that we meet. And if that person is, you know, uh, uh, uh, acting a particular way, we tend to kind of, uh, cement in our minds. This is, this is what happens with, uh, with first impressions. Why first impressions are such a big deal. Psychologically speaking, we cement in our minds, you know, what we believe about that person very quickly. We imagine that if they are, uh, you know, making a certain amount of money that they will continue making that same amount of money. We imagine that if they are, you know, if they act a particular way in a, in a meeting that that is kind of the way they act in all meetings, right? Uh, this kind of relates to fundamental attribution error where we, we attribute somebody's, you know, failure or, or their behavior in a moment to something about them at a very personal level, right? Um, that it's an attribute of who, of who they are, you know, rather than, you know, uh, an expression of a response to this, to this particular environment or some kind of, uh, influence on them, in that moment. So, so this is all of this kind of comes together because, and, and kind of the underlying theory here is we need to simplify the world. Our brains are optimizing all the time, right? We want to kind of essentially cash the world. We, we want to find a picture that makes sense. And then we operate with that picture in mind. Much of this actually tends to work decently. Well, a lot of the things that we encounter do tend to stay static. Right. Um, you know, if, if we were to think about, uh, and I started to mention the AI model, um, growth earlier. So we imagine that an AI models will continue growing. And our default assumption is that the AI models will grow at a, uh, at a relatively linear rate. But all of the research related to agent, uh, um, you know, effectiveness shows that agents are doubling their effectiveness. Approximately doubling their effectiveness at around a hundred day rate. Right. Uh, and this is, you can go, uh, you know, look, look this data up. Even if I'm off by some percentage, the curve is still exponential. So what this means is, uh, the, the rate at which these things are, are improving, right? The agents that we use every day, the rate that they're improving is itself increasing. So what this means is, this is not intuitive. It's hard for us to, to imagine what this looks like because it's an exponential nonlinear growth curve, uh, to, to be able to understand this, we would have to have experienced exponential growth in other forms, you know, parts of our life. Uh, software engineers have kind of, uh, uh, language for how things grow nonlinearly or, uh, sublinearly, right? Some other things in our lives, we have to understand sublinear linearly. And, uh, for what it's worth, most of these things don't have simple, um, um, growth curves that will, that will last forever. In other words, uh, the, the agent growth weight rate or, um, you know, improvement on particular types of solutions will eventually likely have some kind of diminishing return. Um, that's not a given necessarily, but, you know, it's, it's a plausible idea that we will go from exponential to something that looks like a, looks like linear for a little while, and then it will move towards, uh, something more logarithmic, right? In other words, it's curving off, it's growing less and less quickly. Uh, it's improving less and less quickly. And so, this is counterintuitive because, uh, not necessarily because these things don't happen in nature. They do, right? We, we know that exponential, uh, models exist everywhere. Linear models, uh, exist in a lot of places and logarithmic models exist in a lot of places. Some simple examples we can find in physics as you, for example, uh, um, sound decibel rating, decibel rating is an exponential, uh, uh, uh, growth rate. In other words, the closer you are to something, the louder that thing is, but not linearly. This is why you can be right up next to a speaker or 10 feet away from a speaker or five feet away from a speaker. And, uh, the rate at which the, the sound increases in volume, your, your perceived volume, uh, five, you know, uh, one to five is going to be more of a change than five to 10. Right. Um, I think I got that right. So hopefully you understand the point. The point is that it's not a linear change, right? Um, heat or any kind of particle, uh, dispersion, dispersion models kind of follow a similar, similar pattern. If you were to put your hand 10 feet from a fire, uh, and then put your hand five feet from a fire and then put your hand one foot from the fire, you didn't go from, you know, 10 to one in, in, in kind of linear succession, it's going to get drastically hotter as you get closer to the fire, right? So these, these things do exist and we can gain some intuition about how that works just in our natural world. But most of our social relationships and people tend to, tend not to follow those exponential patterns. And so we have a perception that's shaped largely by our social environment. People tend to change at different rates and mostly linear rates, right? If they're going to change at all. And so a lot of the work that we've done, uh, uh, you know, in, in human history, um, if you plant a seed, you're going to get a plant. Sometimes it grows a little faster than that because the plants themselves will spread. But overall, the way that our minds, uh, uh, have, have kind of adapted over the course of human history has largely been, you know, using linear models to, to make sense of the world. So linear and static have served us well. Overall, uh, um, you know, these, these things can, can definitely lead us down the wrong path, right? People can change for example. And so our static thinking, uh, is, is, will lead us woefully, uh, in the wrong direction. Um, and of course, you know, as we already mentioned, lots of things do not follow linear models, even though our, our default mode or, or like kind of automatic thinking will be in linear. Um, this is also true, you know, if you think about your growth in your, in your, in your job, most people kind of assume that we will have linear growth in their salary or, or close to linear growth in their salary, living wage increase, for example. And this isn't necessarily true. And it certainly isn't true as you get later in your career. This model is something that we set up in our minds, but it may be bad expectations. Maybe we should be thinking a completely different way at least about the normal kind of increases that you would expect in the course of your career. And similarly, it's hard to wrap our minds around situations where people have exponential growth income, right? That doesn't seem to really translate to our minds because we're so used to thinking in terms of stepwise growth. Okay. The third and final, I know we said we're going to do two total and we ended up doing kind of those two that fit each other or model each other a little bit in terms of like static or unchanging rates. The third is the concept of the income. And I think that's a really good idea. I think that's a really good idea. The fourth is the concept of infinite growth, infinite stretching. However you want to think about this, we don't really think about things in discrete terms very well. So for example, we don't often think about our careers in terms of architecting them in stages and finalizing our careers. We tend to only think about that as those stages approach, right? So it's very hard for our brains to wrap around that. And so I think that's a really good idea. around the concept of end of career and how we plan for that. Now we do things that are more or less automatic or cultural, like investing in our investing in, in a retirement account, for example. This is something that is a cultural practice. A lot of people do that and it works well for, for that kind of almost accidental discrete planning, but it doesn't necessarily work. Well, or we don't necessarily do this naturally in other areas, a good place where this shows up. And this kind of ties a lot of these things together. If you are an engineering leader, if you're listening to this podcast, it's a good chance that you are. Engineering leaders tend to be responsible for setting direction for their engineering department, right? Or for their team, or, you know, maybe for some sub function of the engineering org. And a lot of the time, this setting direction feels like a huge ambiguous task, right? And so we end up doing, one of two things is very common. And I think is a common mistake pattern or anti-pattern that you, that you can avoid if you're, you're an engineering leader. One is setting a forever vision. And this is that kind of non-discrete or infinite thinking that you're going to have some perfect state that you're engineering, and that's what the engineering org should be shooting after. And so this target state that you're shooting for is unbounded, right? You're imagining that it is, you know, essentially it's perfection, right? This is, you know, if, if every, if we had all the resources in the world, or if, you know, if we were running perfectly, then this is what it would look like, right? And we'll talk about why there's that's problematic in a minute. The other thing that can happen is we think very linearly about the planning. We'll only do the thing that's directly in front of us to do. And we don't take a step back and think about, you know, a step ahead, right? So the way we think about, you know, target state, if we snap to one or the other of those, a lot of times we miss the most productive thing, which is somewhere in the middle. It's somewhere in the middle, right? The most productive thing for most, you know, vision casting for an engineering department is to think about the most productive thing for most vision casting for an engineering department. Things on the months to years scale. And that's really counterintuitive. It requires us to create a more concrete plan. It's a little more difficult to think in terms of this concrete scale or this time, concrete time horizon. Because it means that we have to say no to a bunch of stuff, right? And that we have to kind of put the pieces together in a discreet fashion. This isn't, this isn't intuitive to us. We also have to think, you know, beyond this kind of linear stepwise and beyond our snapshot as well. So a lot of these things show up because our default modes of thinking tend to serve us poorly when we're trying to build a function, build an engineering department. So these three, these three forms of thinking are the default mode and your job as an engineering.