Every once in a while, I come across a small piece of advice that feels like pure wisdom. It cuts to the core in some way that is almost universally applicable in my life and career. In today's episode, I'm going to share one of those pieces of advice, and we'll probably do more of these style of episodes in the future. This is kind of the core of what we started doing at the very beginning of this show. In today's piece of advice, I learned from two different people. One was Adam Savage. You might know him from the show Mythbusters. And the other person was CGP Grey, who is a YouTuber who is also a host on the podcast Cortex. And I've combined these two concepts. into a single concept for me that kind of plays both sides of the concept. The first is the advice that I heard from Adam Savage, which is, the first one is always a throwaway. And the second is the advice I heard on Cortex, which is, two is one and one is none. You may have heard this saying before, but that was the first place that I had heard it. And really the core insight here is that the amount of flexibility or the amount of freedom that you receive, by having a backup, in other words, having the opportunity to fail at your first shot, is worth far more than the cost of the backup. Let me say this one more time. Having redundancy so that you always have one extra, or some number of extras, but usually it is one, is typically worth far more than the cost of going to that redundancy. Specifically, looking at the cost of, if you do not have redundancy, multiplied by the likelihood of failure. If you are risk averse, you can add a modifier into this utility calculation. And the spoiler alert here is that, generally speaking, most people are risk averse, and most of the time having a backup far exceeds the utility function value. Now I'm not asking you to go and do the math. In fact, that's something I would rarely do. It's about the principle, of backups. And you can apply this in a lot of different scenarios. Obviously, the most clear scenario is to have a backup server, running your application, so that if you have one fail, you can easily switch over to the other. In fact, this is one of the clearest and most obvious value propositions, of having your stuff hosted in the cloud. But you can apply this to, pretty much any other scenario you can imagine. Say you're traveling out of town for an on-site interview. Well, it probably makes sense to have at least two outfits that you would wear, to the interview itself. Instead of just having one. It's possible that that one outfit gets lost. Let's say your bag gets lost. Then you start thinking about, okay, what does it mean to have two? One is in my carry-on and one is in my luggage. That way I've basically reduced the likelihood that I'm going to lose my outfit to nearly zero. But this also provides fringe benefits, or kind of the overflow benefits of peace of mind. You have the option, of choosing one outfit or the other, on the day of. Maybe one outfit doesn't feel quite right, and you have the second outfit to go with. Now I must take a pause and confess that I never thought I'd be talking about, outfits, when we started this podcast. But this principle applies so universally, that you can think about it both in the small and in the big ways. Another simple example is having two charging cables for your phone or your computer. If one charging cable goes out, the cost of the second charging cable, which is probably less than $10, is almost definitely a more acceptable cost than not being able to contact somebody at a critical moment, or even worse, in an emergency. But it's not just about planning for the possibility of a failure, it's also planning for the probability of a failure. When you're trying something for the first time, you should probably have the extra materials. For example, I'm an amateur woodworker. This past fall, I've been working at a factory, This past weekend, I built a little toy catapult for my kids to play with in the backyard. I knew that I would need extra wood in case I ruined some of the wood that I was already using. And so if I didn't have the extra wood, I could have easily caused the project to be essentially impossible to do over the weekend. Having to take a trip to the store because I ruined the first set of wood is much more expensive than just buying the extra wood in the first place. Additionally, many times, and this applies in other cases, not just to wood, many times these resources that you are doubling up on are fungible. In other words, that wood, well, I can use it in a later project. And by the way, as it turns out, I did end up needing at least one extra piece of wood for the little catapult project. If you are planning for your team to go into uncharted territory, you should be thinking about this first option being a failure. This is one reason why... draftsmen tend to use pencils. They expect that they're going to have to erase, perhaps rip the sheet off the board and start over again. And so to sum up this advice into a single point of learning, always leave margin for extra. Always leave margin for extra. Very often that extra is needed. And even if it's not, usually it carries over to the next project anyway. Thanks so much for listening to today's episode of Developer Tea. Huge credit, of course, to the source material here, Adam Savage and CGP Grey on Cortex. If you enjoyed this short piece of advice, of wisdom, I encourage you to subscribe to this podcast in whatever podcasting app you're currently using. Secondly, if you want to discuss this concept and other concepts that we've talked about on the show in the over 1100 episodes that we have, join the Developer Tea Discord community at developertea.com. That's 100% free. We talk about this stuff all the time. You can talk about your career. You can ask questions. You can ask for advice. I'm in there. There's a bunch of other software engineers. And also for what it's worth, not software engineers that are a part of that community. That's developertea.com slash discord. Thanks so much for listening. And until next time, enjoy your tea.