Members only: "Won't Fix" self help
Every major self-help framework of the last two decades falls into one of two camps.
- Camp one is Stoic Acceptance: your problems are features, not bugs, and the path to contentment runs through radical non-resistance.
- Camp two is Relentless Optimization: your problems are solvable if you wake up at 4:30 AM, track your macros, journal with intention, and subscribe to the right Substack.
You have Marcus Aurelius on one side, Tony Robbins on the other, and a $13.4 billion personal development industry filling the gap between them with courses and coaching programs and hardcover books that all say some version of the same thing: you can // should // must be fixed.
I'd like to propose a third option: the reasonable // rational recognition that most of your personal flaws are "Won't Fix" bugs, and the single most productive thing you can do about them is stop trying to patch them.
The "Won't Fix" resolution
Tagging a bug "Won't Fix" doesn't mean it isn't real and it doesn't mean nobody noticed; it means the cost of fixing it exceeds the benefit, or the fix would introduce worse instabilities elsewhere, or the system has already built so many dependencies around the bug that it's become, functionally, a feature. Every codebase of sufficient age accumulates these. They're documented, acknowledged, and largely left alone so the engineers can go build something useful.
Note (and this is important): you are a codebase of sufficient age.
The self-help industry's entire business model depends on convincing you that every single bug in your system is fixable, that with the right framework, the right habits, the right coach, you'll finally refactor yourself into a clean, well-architected human being. But how many of your core personality traits have actually changed in the last decade?
The honest answer, for most people, is...very fucking few.
Why refactoring yourself is a terrible use of resources
Fred Brooks observed that when engineers build the second version of a system, they tend to over-design it, cramming in every feature and fix they wished they'd included the first time. The result is almost always bloated, late, and worse than what it replaced. The lesson developers have extracted from fifty years of living with Brooks's observation is simple: don't rewrite from scratch, and work with what you have.
Self-improvement culture is a perpetual second-system rewrite of the self. You're constantly trying to architect Human 2.0, the version of you that's disciplined and calm and focused and doesn't check their phone 96 times a day (which is, by the way, the actual average for American adults, according to Asurion's widely cited research). But Human 2.0 never ships. You keep accumulating half-finished refactors and abandoned meditation streaks alongside a growing sense that something is fundamentally wrong with your willpower.
The alternative is the wrapper pattern. When you have a piece of legacy code that works but has an ugly interface, you don't rewrite it. You write a thin layer around it, a wrapper, that presents a clean interface to the rest of the system while leaving the messy internals untouched. The legacy code keeps doing what it always did, and the wrapper translates between the old system and the new requirements.
What wrappers look like in practice
The Acceptance camp says: release your attachment to punctuality. The Optimization camp says: build a 47-step morning routine with buffer time calculated to the minute. Won't Fix says: you're going to be late, so build a wrapper. Tell people 2 PM when you mean 2:30. Set your clocks ahead. Automate your calendar reminders to fire 15 minutes earlier than the defaults. You haven't changed yourself, but you've written an adapter layer between your actual personality and the world's expectations.
Epictetus, who spent years as a slave in Rome before gaining his freedom and building one of the most influential schools of Stoic philosophy, would probably say that this approach surrenders moral agency. You're supposed to become virtuous, not fake it with systems. And there's something to that objection.
But Epictetus also spent most of his philosophy drawing sharp lines between what you can and can't control. Is your core temperament something you control? The Big Five personality traits (openness, conscientiousness, extraversion, agreeableness, neuroticism) show remarkable stability across adult lifespans, according to decades of longitudinal research in personalty psychology. You can nudge them, but you can't overhaul them.
If Epictetus were working in DevOps, I suspect he'd be a wrapper advocate.
Giving up correctly is its own liberation
In Kazuo Ishiguro's The Remains of the Day Stevens, the butler, reflects on the decades he spent perfecting his professional dignity at the expense of, well, everthing else. His entire life was a refactoring project: eliminate the personal, optimize for service, become the ideal version of what a butler should be. By the end of the novel he's technically excellent and profoundly diminished. He optimized the wrong thing for forty years because he never stopped to ask whether the specification itself was flawed.
Won't Fix is the practice of questioning the specification. Most of the things you're trying to fix about yourself are only problems relative to some imagined ideal of a person you were never going to be. Your distractibility is a bug in the "focused knowledge worker" spec but might be a feature in the "person who notices interesting things and connects them unexpectedly" spec. Your sensitivity and your stubbornness, your tendency to monologue about niche topics at parties: all Won't Fix, and all load-bearing, and all probably okay in the big, heat-death-of-the-universe scheme of all things.
Stop trying to ship Human 2.0. Tag the bugs, write the wrappers, and get back to building something worth building. The most productive version of you probably looks a lot like the current version of you, plus a few well-placed adapter patterns and minus about thirty self-help books worth of guilt about not being somone else.