What color is the sky? If you answered blue, then you're right, sort of. That's kind of what we're going to be talking about today. My name is Jonathan Cutrell, and this is Developer Tea. And no, we're not going to be talking about the science of the color of the sky, but rather, the question itself, and how quickly you came to an answer. How is that? Is there an underlying mechanism that we can understand better about our brains and then use to our advantage that this question kind of outlines? And this isn't the only question that we do this with. This isn't the only question that we have a very quick answer for. So we're going to talk about some of that today. My goal here on this show on Developer Tea is not just to be the host of a podcast. Instead my goal is to coach you, to coach you through your career, to hope you become a better developer. And really, I'm learning as I go just like you are. So my hope is not only to coach you, but to stand beside you and help you fight through the things that you're fighting through as developer. So if you're listening to this show, then hopefully you are willing to be coached. Hopefully you're willing to work and to put energy into becoming better. If you're here for a quick fix, if you're here for hacking your life in some way to jump ahead without putting in the hard work, well, this probably isn't the place for you. If you're here to hack your life and jump ahead and also put in the hard work, this is exactly the place for you. That's exactly what we do here. I don't try to help you be lazy or cut out work where work is necessary. Instead I try to help you work on the things that matter the most and eliminate the things that matter the least. So I hope you can adopt that perspective of this show that really it's just a coaching session between you and me over a cup of tea. That's what I hope for this show. And I also hope that you will take the opportunity to really view this as a give and take. I really would love to hear your feedback. I'd love to hear your questions. Any comments you have on these coaching sessions. These podcast episodes and how they apply to your career and to your life. You can reach out to me at developerateatgmail.com. Let's jump into this discussion on question substitution. Your brain does a really good job of this. Substituting one question for another is something we automatically do to make hard questions easier to answer with relative accuracy. So let's think about that question that we asked at the beginning of the show. That color is the sky. And the funny thing is our brains don't automatically call the memory our science class where we learned that the sky's color changes because the sky itself isn't really what has the color. It's actually the refraction of light by particles in the atmosphere. And this is the basic way of understanding it from a scientific perspective. And yet we don't answer the question what color is the sky by outlining even the most basic of scientific answers. Instead we jump directly to blue. Or if you're standing outside you may look up at the sky and answer from that perspective. So really what's happening is you're substituting a different question for the original question. Instead of answering what color is the sky. Polistically you're answering the question what color do we normally perceive the sky's refraction to be or perhaps what color am I currently perceiving the sky's refraction to be an interesting additional note here is that this particular question is easier to answer and it's harder to ask. It's a little bit more wordy. It's a little bit more thought out. And the sky isn't the only thing that we substitute questions for. The sky's colors not the only thing we substitute questions for. We have also get ourselves these shortcuts in harder concrete problems like for example longer multiplication problems. Even with its relative simplicity and the scale of math. If I were to ask you the question what is 254 times 763 you won't immediately know that the answer is 193,802 you'll probably do what I did. The way we answer this question is in and of itself a substitution. Instead of actually finding 254 objects let's say marbles and replicating them 763 times and creating some grid and counting up the result. Instead of doing that we type in symbols on our computer's calculator and ask the question what does the calculator tell me when I type in a representation of this question and hit enter. That's the real question that we're substituting even though that question is representative of the original question. While this automatic substitution is extremely helpful in many scenarios it can also be detrimental and others and furthermore we can use this concept more intentionally to help us create more space for focus and open up our minds to new perspectives and possibilities. We'll talk about these characteristics right after we talk about today's sponsor pusher. Today's sponsor pusher allows you to stop thinking about whether or not you can build features that are real time into your application and instead start thinking about which features you want that are real time. Users like Adaptive UI and AB testing that can adapt on the fly for your application based on data that's coming from other users who are currently using that application. Features like chat. All of these features are made much simpler by pusher. Pusher has a hosted API that makes it simple to add these real time features to your web and your mobile apps in minutes and you can scale up massively with pusher's API. Pusher allows you to use the power of web sockets to connect from your server to the clients that are using your application. It doesn't matter what language or framework you are used to, they will have a library that works for you. They have over 150,000 developers who are using their service and big companies like GitHub, Intercom, MailChamp and The New York Times rely on pusher to build their real time features. To go and check it out, spec.fm slash pusher, you can get started today with a free plan at spec.fm slash pusher. Thank you again to pusher for sponsoring today's episode of Developer Tea. Let's talk about the most important characteristics of a substitute question. The most important characteristics of a substitute question. Really there's one characteristic here and a lot of sub-characteristics where things to consider about this characteristic that is important and that is whether or not that substitute question can accurately and confidently answer the question it is substituting for. Let me say that again. Can your substitute question accurately represent an answer for the question it is substituting for? In our previous example, we talked about the color of the sky and the color of the sky is entirely dependent on your perspective. Right now I'm looking at the sky out my window and it's largely gray and kind of a hazy blue. It's not really what we imagine when we think about that simple what color is the sky. Sky blue is a color that we have kind of ingrained in our minds. It comes from some cultural aspects of carrying that idea forward of what color is the sky. Perhaps it comes from the most common representation of the sky. Usually the sky is not cloudy in most parts of the world. In the main portions where the world is populated, when we look up we see the color blue. The substitution of the question what color is the sky for what color do I perceive the sky to be or perhaps the better question would be what color do most people perceive the sky to be on a regular basis. That kind of question, we have a pretty good way of determining the answer for. So the substitute question in that scenario is relatively good. However, take a question that has a lot more information to consider. For example, is milk good for you or is bread good for you? There's a lot of research that you may call to mind. There's a lot of news articles that you may call to mind as well. More information than you can really digest and immediately spit on an answer for. You're going to substitute that question automatically with something that you believe based on a construction of the information you've received. Though I don't have the actual answers to these questions, the most likely answer is it depends. It depends on perhaps you are particularly lactose intolerant. Well then milk probably isn't the best choice for you. Of course, this doesn't just apply to food or nutrition. It also applies to the way we perceive other people, the way we judge other people and their abilities. We ask, is this person going to be good for this project? And really we're trying to interpolate the information we have from the past and project something into the future about their abilities, about their behaviors, about their way that they deal with stress, for example. We may use information to create a simplified answer and ultimately we're creating a non-explicit connection to a substitute question. In other words, is this person going to be reliable on this project? We may take from our information about that person who, let's say in this scenario, they are on time regularly. We may take from their on-timeness to extrapolate that concept into them being a reliable person in every part of their life. And so the actual question we are answering is, are they on time regularly? And that substitute question stands in place for the larger question, will this person be reliable on this project? Of course on the flip side of the scenario, if you're not the one asking the question, but rather the one who the question is being asked about, then you know that these substitutions could occur. You know that for example, if somebody sees you being late, well they may substitute that lateness for, is this person lazy? You may not be lazy at all. It may be that you face traffic on a regular basis. And so you often are late because traffic is hard to predict based on where you live. Now what are you supposed to do in that scenario is entirely up to you. The point of this discussion, the point of today's episode, is to help you become more aware of this reality, help you become more aware of the question substitution. But I also want to help you pick better substitute questions on purpose intentionally, rather than the ones that you're doing subconsciously, you can use this as a tool. Right? So what I want you to start doing is thinking in degrees of confidence in these substitute questions. This will likely be covered in another separate episode. We've discussed degrees of confidence and similar concepts before when we talked about heuristics and when we talked about estimation. But here's the idea in almost all scenarios substituting a new question for a more difficult question substituting an easier to answer question that will decrease the degree of confidence and accuracy in the answer. Think about that for a second. Anytime you substitute an easier question for a harder question to answer, then you're sacrificing some degree of confidence that that answer is actually representative of the question it is substituting. Now this isn't always true. For example, in the multiplication problem, I actually trust a calculator much more than I trust my own long form math abilities. But this is especially true with more complex questions that rely on context, subjective answers, research rather than a one-dimensional easy to determine fact like for example, lateness. So what I want you to do is start thinking in degrees of confidence, determine especially for intentional question substitutions, your degree of confidence, and you can use a scoring system if you want to, one to ten, how confident are you, your degree of confidence that the new question provides an accurate and representative answer for the original question. Of course, you want to find substitutions that have high degrees of confidence. Interestingly, we can be very wrong about our degrees of confidence in a given answer. Consider the often repeated phrase, correlation is not causation. When we rely purely on correlation as a heuristic forecazation, there's no guarantee that the substitution of one side of the correlated factors for the other is a valid move. I'll give you a fake scenario. Joe, the scientist, is trying to understand what factors cause sunburn in young children. And he does some studies, he collects some data, and he analyzes that data only to find out that there's a high correlation with ice cream consumption per capita and the reported rate of sunburns. Now we can look at that and hopefully you're already starting to see the answer, but we could look at that and tell children that they no longer allow to have ice cream because ice cream certainly causes sunburns, or we can take a step back and analyze that data and recognize that the correlation between sunburns and ice cream is really nothing to do with each other, but rather they both have something to do with summer. And children are more often outside in the summer and they are more likely to have ice cream because it's hot outside. So of course a non-useful substitution in this scenario is this child likely to get a sunburn if you were to substitute how often does the child eat ice cream, then it would yield a poor result even though the data may say otherwise. So we have to be vigilantly aware that first of all we are making these substitutions all the time automatically and secondly we have to be vigilantly aware that when we do these intentionally, when we substitute questions for other questions in order to come to an answer faster that we very often can be wrong. We very often trust things that we shouldn't trust. Thank you so much for listening to today's episode of Developer Tea. I hope you've enjoyed this episode. I hope that you feel the same momentum from this coaching that thousands of others just like you are feeling. They've subscribed to the show. I hope you will subscribe to the show as well so you don't miss out on future episodes of Developer Tea. We talk about stuff like this. All the time on the show we talk about psychology, about focus. I invite you to join us and stick around. This show would not be possible or even important without you listening to it. So thank you so much for listening to the show and I hope that you will reach out. You can email me at Developer Tea at gmail.com. Thank you so much for listening to Developer Tea and another huge thank you to pusher. Our sponsor for today's episode with pusher, you can start putting real-time features into your web and mobile apps in just a few minutes. They have a hosted API and libraries for pretty much every language. Go and check it out. Spec.fm slash pusher. You can get started for free today. Thank you so much for listening to today's episode of Developer Tea. Until next time, enjoy your tea.