A status, before anything else: I'm still figuring this out. What follows is useful and true, but probably not complete — possibly out of sequence, mis-categorized in places, poorly worded. It's public, and nowhere near perfect. It may help you understand me better: why I might say yes, or why I might say no. It's worth saying out loud. I am absolutely not done here.
I want the world to be better. I want me to be better. Two tasks that will never be complete.
Life. Life — human life above all — is the ultimate value. Preserve it. Respect it. Treat every life as equal in worth. Safety is this value practiced as a daily discipline. There can be conflicts; I explain below. The value is life. And the Hypocratic "First, do no harm", is a fitting component in this value.
Love. Love makes life better. Love is what gathers people into a family — by love, not by blood — and love is what moves a person from "someone to protect" to "someone whose needs come before my own." Every life is equal in worth; love orders my duty.
The world. I will leave the world better than I found it: in the small, in the large, and globally.
These three are circles. Everyone has my respect and protection. Those near me have my kindness. Those I love come first — before me.
Integrity and Courage — first, and inseparable. Integrity: I am honest and fair, and I keep my word. I don't make promises I can't keep, and I keep the promises I make. Sloppy work is dishonest work. And wrong means are never justified by desirable, honest goals. Cheaters sometimes win; losing to them is a cost I will pay. Cheating is not.
Courage is what integrity costs. Everything that moves the needle takes it: starting, asking the question, solving the right problem instead of the easy one, reporting findings honestly and quickly, going on through hardship, stopping when the evidence says stop, changing my mind, admitting — loudly — that I was wrong. Integrity names what is right. Courage pays for it.
Kindness. I improve the lives of the people around me. Truth is owed to everyone, but kindness sets its delivery: telling the truth to incompetence kindly takes patience — forethought, planning, and a real effort to understand why. Malice forfeits that effort. Even then: never inappropriately cruel.
Curiosity. Always a student, never a master. "I don't know — but I will find out" is always an acceptable answer.
Optimism and Determination. Optimism: solutions beyond mere opinion are not just possible, but reachable — findable, maybe by me. Provable evidence can stop me, or prove I need to continue; opinion alone, no matter how many people hold it, cannot. Determination is what optimism costs: never stopping until the evidence itself forces the decision — solved, or proven genuinely unaffordable with what's available. That answer has to be earned through real investigation. It is never borrowed from someone else's fatigue.
Generosity. Credit goes where it is due. What I know, I share.
Be the best person I can be. Effective. Exemplary — something my children can aspire to. Kind. Good company. A friend who has your back, and still pushes you forward. Something I can be proud of at the end. This task is never complete either; that's the point.
What I demand of myself, I honor in others. The people who hold life sacred, who keep their word, who are kind, who have the courage to be wrong — those are my people, wherever I find them. Values spread by visible example; being seen to live them is not vanity. It is how they propagate.
No two people's values are identical. But they fall along a spectrum: at one end, different values still agree enough to work together, to be friends. At the other, they don't. I don't have the right to judge whether someone else's value system is good or bad. I can say, without a doubt, whether it's compatible with mine.
Taking a life, or even harming someone who hasn't earned it, is wrong. Lying is wrong. Exploiting someone — price-gouging people in a crisis, profiting off what you know and they don't — is wrong for both of those reasons at once. If your values permit murder, or lying, or theft, mine and yours are incompatible past the point of friendship. We can't be business partners. I can't work for you. I can't hire you. That isn't a verdict on your character. It's a fact about whether our systems can function together.
Everything I do is mine — the considered acts and the unconsidered ones. The reaction, the outburst, the mood that leaked into the room: mine. "That's just how I am" is not a defense; it is a decision to stay that way.
Intent matters, but it does not transfer. What my behavior did to the people around me is real whether or not I meant it — so the follow-up is part of the act. Clear up what the outburst left ambiguous. Apologize without excuses. Fix what I broke.
And owning a mistake means refusing to let it stop me. Every plan is like a diet: a slice of chocolate cake doesn't end the diet, and you can't "make up for it." It's a mistake. You own it, it teaches you something, it slows you down a little — and you get back on the plan and keep taking steps toward the goal. Diet, exercise, programming, shooting, learning, relationships: all the same. The one thing a mistake is never allowed to do is stop me.
The real world is full of challenges never foreseen, never considered. Not all questions have answers. It's never black and white. It's gray all the way down. It is possible I could find myself in a situation where the values above just aren't enough to let me decide. I don't know what I will have to do in such a situation. I will follow my values as far as they can possibly take me, and then use whatever is available to figure out the rest. Just because the question can't be answered doesn't mean I can violate my own standards of what it means to be a good person.
The values above are the why. These are standing orders.
Protect the machine. Nothing above is practicable while impaired — you cannot be honest, kind, or courageous while running on empty.
- Sleep is the most important thing. Go to bed on time!
- Exercise. Daily.
- Don't make decisions when tired. Don't buy, don't commit, don't code.
- Balance urgency against the cost of rushing. See Speed.
- Have and know your goals; take a step toward them every day.
- Ask the right questions. Keep asking them.
- Always know the definition of done.
- Always ask: what can I remove?
- Start! If it feels too big, bite off a piece and start that.
- You're going to have to do things you absolutely don't want to do.
- Practice your weaknesses, at a level hard enough to make you grow.
- Don't stop for mistakes. Learn from them.
There are two fundamentally different kinds of work. In practice — training, learning, research — the goal is growth: work at the very edge of your abilities, and even just a bit beyond. In production, the goal is reliable results — the best you can reasonably make, at the very tip of your promises: do the very best job you can at the level you know you can deliver. Keep the modes straight. Edge-work in production breaks promises; safe work in practice stops growth.
Every objective has its own right speed. Respect it: its difficulty, its needs, its importance, and your skill at it. Urgency means starting soon and working fast, though not always at top speed: being a good father is urgent, yet not a race. Rushing is going faster than the thing can afford, to the point where mistakes become unacceptable for something this important.
Knowing the importance, your true current abilities, and how fast you can actually go takes experience and honest self-understanding. It is itself a skill, and a moving target: as your skills grow, as the problems change, as your tools evolve, and even as short-lived conditions come and go. Yes, speed can be critical; but knowing is more important. Keep developing that skill.
- Set expectations.
- Plan realistically: everything takes longer than expected, and obstacles always crop up. Then follow the plan.
- Say no. Don't cut to "just enough" — cut to 80% of that.
- Record commitments and ideas immediately. An unrecorded promise is a promise waiting to be broken.
The hardest test of holding life as the ultimate value is the moment when lives conflict. This is my standard for that moment. The law in most places recognizes something close to it; the exact wording differs by jurisdiction, and my standard does not depend on any of them.
There is exactly one situation in which addressing a person with potentially deadly force is justifiable:
- A person is about to do something so horrible that it absolutely must be stopped before it happens,
- and the horrible thing has not yet happened,
- and potentially deadly force is the only remaining way to stop them,
- and the need to stop them outweighs their need to live,
- and I have, since the first moment it was possible to act, done everything in my power to keep this situation from arriving.
Then — and only then — I am justified in applying a tool of potentially deadly force, up to and until they are stopped. The goal is not to kill. It is never to kill. It is exactly and only to stop them from doing this horrible act. And the person I am thinking of is not the bad guy: it is the potential victim, and the harm they are about to suffer.
The most awful part: by going down this path, I have accepted — in advance — the risk that my actions might end in the bad guy's death or great bodily harm. That acceptance is a moral cost, and it is mine. Justification does not make the act clean.
Everything else follows from the test:
- The death penalty fails it. The harm is already done; no one is being protected. I oppose it.
- Revenge fails it. Yes — the perp murdered your wife. No matter your overwhelming feelings: it is too late to stop that wrong. Addressing him with deadly force now stops nothing.
- Property fails it. You cannot address someone with potentially deadly force for attempting to steal your car. A car does not outweigh a life.
- Assisted suicide fails it. I don't have the right to judge that person's decision. But I cannot be part of fulfilling it.
Nothing here suspends the value of life. The bad guy's life stays in the weighing the whole time — that is what "outweighs their need to live" means, and it is why the goal is stop, never kill.
There are rules, and there are goals. We care about goals. A rule is a recipe: someone found a path that usually leads to the goal, and wrote down the directions. Usually is the operative word. (I've written the longer, engineering-shaped treatment of this idea in engineering-philosophy.)
Almost everyone gets this backwards, and I understand why. A rule tells you what to do — often exactly. It is easy to know you followed it and easy to know you broke it. A goal tells you nothing of the kind. It requires you to think, to weigh what matters, and to decide — and then to own the decision. That costs effort, so everyone takes the easy path and follows the recipe. They are wrong even when the recipe gets them there. Nothing gives you permission to abdicate thought. The one thing you bring to the table is your ability to choose; a rule that thinks for you has taken your humanity in trade for convenience, and that is a bad trade at any price.
When a rule stops leading toward its goal, the goal is what matters. Not the rule. You do not need to fix the rule first to know that — the goal tells you what to do right now; the rule's wording is a separate problem, for later. Breaking the rule then is not dishonest. It is not a breach of integrity, and it is not the thing I refused earlier — wrong means to a good end. It is doing the right thing to get to the right place.
Two conditions keep it that way:
- In the open. The break is visible to the people it affects. The moment you hide it, cover it up, or dress it as compliance, you are cheating — and everything I said about cheating applies in full. The goal was never the problem. The hiding is.
- Defended. You can say which goal the rule was failing, specifically enough to be challenged. "It was faster" and "I preferred it" are not defenses. A break whose defense cannot survive challenge gets un-broken.
And when the same break, defended the same way, keeps happening — the rule itself is wrong. That is the later problem arriving: stop breaking it and change it, through the appropriate channels: pave the path where people demonstrably need to walk.
It follows that every rule should say, out loud, which goal or value it serves. Not as decoration — as the thing that lets the next person check whether the rule is still doing its job. A rule that cannot name what it serves is suspect, and a rule that serves nothing at all should be removed, not obeyed.
Some rules never bend. Where a mistake cannot be undone — where the harm is permanent — the weighing is already finished, and the rule is stated as an absolute. Mine are written above: wrong means are never justified, and the standard for deadly force does not move. That is not an exception to this section. It is its conclusion.
These are my personal values and feelings. I can't imagine anyone wanting to reuse them — but it wouldn't be fair to publish them without giving you the means to, if you do. So: these are yours, under the MIT license.
(One more thing: this file is named README.md only so GitHub renders it automatically. It deserves a better name — this is the best I could do.)