I'd like to return to optimisation order, the idea that when you're optimising for multiple things, the one that you start with gets the best result. In the Gale-Shapley algorithm, used among other things in matching medical students with hospitals, the students apply in rounds to their preferred hospitals who accept or reject their applications, retracting existing offers if necessary. Although this might seem to put hospitals in the stronger position, in fact this system is known as "applicant-optimal". If there are multiple solutions, students will get the best acceptable match, whereas hospitals will get the worst. The key observation is that a student's preferences are evaluated first (when they apply) and a hospital's preferences are evaluated last (when there's a conflict).
Thinking in terms of optimisation order makes you aware of distinctions that are otherwise invisible. For example, when talking about features to add to some software program. "Oh, can't you just add encryption to it? What's so hard about adding encryption?" On paper, perhaps, nothing. The problem isn't this individual decision, it's the question about where this decision is coming from and whether you make this decision before or after you make others. If your goal is to be a secure messenger first, and an easy-to-use messenger second, then sure. But what if your goal is to be an easy-to-use messenger first and a secure messenger second?
Sure, yes, both, of course we want both. But let's say that you have to make some minor compromises along the way to do that. Maybe adding encryption means you can't store messages on the server, so people who get new computers lose their message history. Maybe adding encryption means the app has to do some verification song and dance every when you add a new contact and it just confuses people. Hell, maybe it just makes the settings page marginally harder to understand. The point is there's always a cost, always some tradeoff where you have to start saying, yes, some of these encryption priorities need to come before your usability priorities. So now you're not easy-to-use first, and someone else who is will beat you if people care more about that.
The difficult thing here is that, while individual actions are visible, optimisation order isn't. Let's say you want the best job you can get. You go through every job ad, rank them carefully, apply in order, prepare well, interview well, and in the end you get your first preference. You win, right? Nope, you lose. The problem is that you interviewed for the first company that posted a job ad. In other words, the companies get the first pass of optimisation when they decide whether to hire and what to hire for. You then get the second pass when you decide which of those companies is your favourite. If you want the best possible outcome, you need to start with your list of dream jobs without considering whether they're hiring, and then work down the list trying to get hired at each one anyway.
This is also part of the reason why sales situations are so easy to get turned around in. Salespeople set the bounds of the decision, they say "we could give you option A with cost W and benefit X, or option B with cost Y and benefit Z", and you get to choose between the options that have already passed the good-for-them filter. If you're good at working that situation, you can get the best option they will offer you, but you'll never get the best option they wouldn't offer you. That would require you to become the salesperson and start trying to convince other people to accept your terms instead.
One last thought on this, which is that, okay, you may end up paying too much for something, or get your 5th most preferential job instead of your 3rd, but that might not be so bad. It's just a TV, or a job, or whatever. But no matter who you are, there's one area where you can't afford to forget about optimisation order: your own decisions about what you want. Are you running your life starting with your preferences, or starting with someone else's and figuring yours out afterwards? Are you trying to find a way to make the world work with how you want to be, or are you taking on the world's expectations and dealing in leftovers?
Who's on first? To my mind, there's only one acceptable answer.