You don’t need a better system. you need less system. if your brain feels “busy” 24/7, it’s probably not burnout — it’s context obesity: too many open loops, too many tools, too many identities pretending to be productivity. Notion, ai, courses, frameworks — they don’t save you. they quietly turn into a museum you maintain at night to feel like you’re trying. this piece is a demolition plan. context obesity you’ll see: where your attention leaks (the four layers of your personal operating system) why you need a minimal stack, not a zoo of tools how to write an honest self.md file — a set of rules for how you treat yourself (and AI) that you can return to anytime where your attention leaks (the four layers of your personal operating system) four layers why you need a minimal stack, not a zoo of tools minimal stack how to write an honest self.md file — a set of rules for how you treat yourself (and AI) that you can return to anytime self.md file in the previous article, we said something simple but annoying: previous article often, you’re not burned out — you’re living with context obesity. context obesity your head is clogged with unfinished tasks, AI keeps pouring more on top, and Notion + Obsidian turn into a second job. after publishing, we dropped the text into Space — our closed community inside AI Mindset — and things got spicy. here are a few quotes (with permission). Space Space AI Mindset AI Mindset “the recommendation ‘don’t keep more than three active projects in your head’ — that’s the hardest task. it’s scary to choose wrong. and those three projects might be imposed by external authorities and lead you somewhere you don’t even want to go.” — Dmitry Kompanets “the recommendation ‘don’t keep more than three active projects in your head’ — that’s the hardest task. it’s scary to choose wrong. and those three projects might be imposed by external authorities and lead you somewhere you don’t even want to go.” — Dmitry Kompanets another participant shared: “i dumped all tasks until the end of the year into Obsidian. asked Cursor to build a day plan. it came out to 29.5 hours of tasks including sleep and rest. i closed Obsidian and realized: first i need to rewrite my system prompt — i don’t have three projects. and yes, that’s… another task i now have to finish 😭” — Ekaterina Gracheva “i dumped all tasks until the end of the year into Obsidian. asked Cursor to build a day plan. it came out to 29.5 hours of tasks including sleep and rest. i closed Obsidian and realized: first i need to rewrite my system prompt — i don’t have three projects. and yes, that’s… another task i now have to finish 😭” 29.5 hours — Ekaterina Gracheva a third added: “start of the day: a mini standup with 3 key goals. end of the day: recap + gratitude. it helps you finish the day with ‘i did good’ instead of ‘another day flew by and i did nothing.’” — Ksenya Tkacheva “start of the day: a mini standup with 3 key goals. end of the day: recap + gratitude. it helps you finish the day with ‘i did good’ instead of ‘another day flew by and i did nothing.’” — Ksenya Tkacheva and then an imaginary Mark Manson in our internal “board of directors” dropped the summary: “honestly? you’re not only suffering from context obesity — you’re suffering from context narcissism. you’re chasing the image of a productive person, not what you actually need. AI and prompts are a curtain hiding the real question: maybe you’re doing the wrong thing entirely?” — Alexander Stashenko “honestly? you’re not only suffering from context obesity — you’re suffering from context narcissism. you’re chasing the image of a productive person, not what you actually need. AI and prompts are a curtain hiding the real question: maybe you’re doing the wrong thing entirely?” context narcissism — Alexander Stashenko this text is about that. not “10 more frameworks.” but the museum of systems, context narcissism, and how to write a living self.md that doesn’t instantly get archived into shame. self.md an evening in the museum of systems imagine: 22:30, you decide to “get your life together.” 22:30 you open Notion. Notion in there: a task database “by the GTD rules” (GTD / Getting Things Done — David Allen’s system: dump everything out of your head into a “trusted external system”) a separate “ideas” database a “goals of the year” dashboard by life areas a task database “by the GTD rules” (GTD / Getting Things Done — David Allen’s system: dump everything out of your head into a “trusted external system”) Getting Things Done a separate “ideas” database a “goals of the year” dashboard by life areas then you switch to Obsidian. Obsidian different folder logic. a “second brain” based on Building a Second Brain (one digital head instead of ten scattered vaults) journals, notes, a link graph that used to feel inspiring — now it just presses on your chest a “second brain” based on Building a Second Brain (one digital head instead of ten scattered vaults) Building a Second Brain journals, notes, a link graph that used to feel inspiring — now it just presses on your chest in downloads there’s a YearCompass PDF — the annual review booklet. you open it, scroll past questions like “how was your year?” and “who do i want to become?”, catch a wave of shame — and close it. YearCompass in Telegram, a habit tracker bot quietly blinks. Telegram little circles: “sleep / sport / meditation” half-empty. your inner voice goes: “yep. another day where i didn’t become the upgraded version of me.” and on top of that: a course called “build your perfect system” you never finished a “20 Notion templates” pack a video titled “how i organized my entire life in one dashboard” a course called “build your perfect system” you never finished a “20 Notion templates” pack a video titled “how i organized my entire life in one dashboard” and now you’ve spent 40 minutes “organizing” — but realistically you’ve just been wandering around your own museum of systems. 40 minutes “organizing” leaving the museum feels scarier than redesigning it one more time. context narcissism: not a diagnosis, a defense mechanism quick clarification: this is not a “you’re all narcissists, go get treated” post. i’m not writing from the outside. i’m writing as a former curator of my own museum — someone who loved the feeling of being “too complex for simple solutions.” loved context narcissism isn’t “you’re vain and bad.” context narcissism it’s a protective mechanism that: is afraid to admit your current setup is physically impossible to carry is ashamed to say “i don’t want to live like this, even if it looks impressive” replaces an honest “no” with a beautiful glass display case:“look how complicated my life is. look how many systems. look how many projects.” is afraid to admit your current setup is physically impossible to carry is ashamed to say “i don’t want to live like this, even if it looks impressive” replaces an honest “no” with a beautiful glass display case:“look how complicated my life is. look how many systems. look how many projects.” and in a world where: Notion setups look like interior design magazines it’s “normal” to have GTD + a second brain + annual reviews + quarterly planning admitting “one system and three projects is enough” feels almost embarrassing… Notion setups look like interior design magazines it’s “normal” to have GTD + a second brain + annual reviews + quarterly planning admitting “one system and three projects is enough” feels almost embarrassing… the museum gives you the soothing feeling of: “at least i’m trying.” but it has a side effect: every new system = a new layer of unfinished expectations. you didn’t just fail a project — you also “failed” your templates, your courses, your inner coaches. context obesity grows. every new system = a new layer of unfinished expectations. you didn’t just fail a project — you also “failed” your templates, your courses, your inner coaches. context obesity grows. it’s not that you “broke your productivity system.” it’s architecture when we looked at people’s tool stacks in the AI Mindset lab, the pattern was almost always the same: 2–3 task systems 2–3 note systems several formats of annual/quarterly reviews lots of guilt, very little sense of trajectory 2–3 task systems 2–3 note systems several formats of annual/quarterly reviews lots of guilt, very little sense of trajectory so we stopped thinking in framework names and started thinking in layers. layers we got four layers: core — habits + execution: how you process inputs and actually take steps file system — where your knowledge, notes, and decisions live compass — who you are right now, and where you’re even going interface — how you look at your life “from above” and reframe it core — habits + execution: how you process inputs and actually take steps core file system — where your knowledge, notes, and decisions live file system compass — who you are right now, and where you’re even going compass interface — how you look at your life “from above” and reframe it interface this architecture isn’t here to “explain everything.” it’s a knife. it lets you cut the museum down instead of expanding it. below is a quick walk through each layer — with questions and practices. layer 1. core: are you living, or endlessly preparing to live? the core answers one question: “i said it — did i do it?” “i said it — did i do it?” realistically, the core should be: one task list (any app, board, notebook — doesn’t matter) 2–3 anchor habits from the Atomic Habits universe: sleep, movement, a screen-free window one task list (any app, board, notebook — doesn’t matter) 2–3 anchor habits from the Atomic Habits universe: sleep, movement, a screen-free window Atomic Habits any sane task system, at the base, is painfully simple: everything in your head drops into an inbox once a day/week you process it each item becomes a concrete next step or an honest “no, not now” everything in your head drops into an inbox once a day/week you process it each item becomes a concrete next step or an honest “no, not now” ask yourself: how many different places do my tasks live right now? which habits am i tracking, but haven’t actually done for months? how many different places do my tasks live right now? which habits am i tracking, but haven’t actually done for months? practice: “minus two trackers” practice: “minus two trackers” for one month: pick one task system you’ll actually live in. disable the rest: hide icons, remove widgets, turn off notifications pick two habits that genuinely change your state (sleep, movement, time without screens) — and delete everything else from the tracker pick one task system you’ll actually live in. disable the rest: hide icons, remove widgets, turn off notifications one pick two habits that genuinely change your state (sleep, movement, time without screens) — and delete everything else from the tracker two layer 2. file system: a second head, or a warehouse of shame? your file system is your digital memory: Obsidian, Notion, Evernote, Google Docs — doesn’t matter what what matters is whether it helps you think and decide, or just reminds you how much you “should read and structure” Obsidian, Notion, Evernote, Google Docs — doesn’t matter what what matters is whether it helps you think and decide, or just reminds you how much you “should read and structure” the “second brain” idea is simple: one extra head where you keep what you actually need for your projects and life areas — not the entire internet. actually need ask: when was the last time i opened a note that changed a decision — not just numbed my guilt because “at least i saved it”? when was the last time i opened a note that changed a decision — not just numbed my guilt because “at least i saved it”? practice: “archive without repentance” practice: “archive without repentance” everything you haven’t touched in 3+ months: tag it archive and remove it from the first screen don’t organize. don’t beautify. don’t swear you’ll return make a deal with yourself: in the active system, only what i actually touch is allowed to live everything you haven’t touched in 3+ months: tag it archive and remove it from the first screen archive don’t organize. don’t beautify. don’t swear you’ll return make a deal with yourself: in the active system, only what i actually touch is allowed to live one month on a “lightweight second head” noticeably melts context fat. context fat. layer 3. compass: who is making decisions — you, or your “internal authorities”? this is where it gets uncomfortable. your compass is: the answer to “who am i right now?” (founder, employee, researcher, parent, a cocktail) the theme of the year: one sentence instead of a goal salad — “year of health,” “year of honest no’s,” “year of leaving the systems museum” a brutal audit of projects: “mine” vs “imposed” the answer to “who am i right now?” (founder, employee, researcher, parent, a cocktail) the theme of the year: one sentence instead of a goal salad — “year of health,” “year of honest no’s,” “year of leaving the systems museum” a brutal audit of projects: “mine” vs “imposed” remember that Space quote: “it’s scary to choose three projects. what if all three are imposed?” “it’s scary to choose three projects. what if all three are imposed?” context narcissism is strongest here: it’s easier to fill out another annual review or draw another mind map than admit half of what you carry isn’t your life — it’s other people’s expectations. quick definition: what i mean by project. project not “reply to a client” and not “make a landing page,” but something that: lives for weeks or months requires recurring attention pulls a chain of tasks and decisions behind it if canceled, changes the plotline of “who i am right now” lives for weeks or months requires recurring attention pulls a chain of tasks and decisions behind it if canceled, changes the plotline of “who i am right now” projects are major storylines. tasks are bricks inside them. practice: “3 projects that are actually mine” practice: “3 projects that are actually mine” write down all current projects: work, side, study, personal label each with one tag: mine — it wouldn’t exist without me, and i genuinely want it to exist imposed — i do it to not disappoint / to fit in / to avoid looking like a loser phantom — i never move it, i just drag it around in lists choose exactly three “mine” projects for the next 3 months everything imposed/phantom becomes either: an honest “i’m not doing this,” or “someday, but not before [specific date]” write down all current projects: work, side, study, personal label each with one tag: mine — it wouldn’t exist without me, and i genuinely want it to exist imposed — i do it to not disappoint / to fit in / to avoid looking like a loser phantom — i never move it, i just drag it around in lists mine — it wouldn’t exist without me, and i genuinely want it to exist imposed — i do it to not disappoint / to fit in / to avoid looking like a loser phantom — i never move it, i just drag it around in lists mine — it wouldn’t exist without me, and i genuinely want it to exist mine imposed — i do it to not disappoint / to fit in / to avoid looking like a loser imposed phantom — i never move it, i just drag it around in lists phantom choose exactly three “mine” projects for the next 3 months three everything imposed/phantom becomes either: an honest “i’m not doing this,” or “someday, but not before [specific date]” an honest “i’m not doing this,” or “someday, but not before [specific date]” an honest “i’m not doing this,” or “someday, but not before [specific date]” [specific date] yes, this hurts more than building a new dashboard. but without this, any self.md becomes just another pretty museum exhibit. self.md layer 4. interface: are you recording life, or just falling asleep and waking up? interface = any way you look at your life “from above”: annual review weekly review a tiny daily ritual annual review weekly review a tiny daily ritual it doesn’t need to look like an Android 2011 launcher stuffed with custom icons. Ksenya’s Space ritual looks like this: morning: mini standup — 3 realistic goals evening: recap — outcomes + gratitude morning: mini standup — 3 realistic goals evening: recap — outcomes + gratitude nothing fancy. but it stops days from dissolving into “i did nothing again.” practice: “i did good” for 7 days practice: “i did good” for 7 days for 7 days: morning: write 3 tasks that realistically fit into today evening: answer 3 questions: what important thing did i do today? what good thing happened, even if the day was hard? what am i grateful for — to myself / to the world — today? morning: write 3 tasks that realistically fit into today morning: write 3 tasks that realistically fit into today evening: answer 3 questions: what important thing did i do today? what good thing happened, even if the day was hard? what am i grateful for — to myself / to the world — today? evening: answer 3 questions: what important thing did i do today? what good thing happened, even if the day was hard? what am i grateful for — to myself / to the world — today? what important thing did i do today? what good thing happened, even if the day was hard? what am i grateful for — to myself / to the world — today? one week later, you’ll notice days start ending — not just collapsing into a gray feed. the minimal stack: an experiment against the museum if i compress everything above into one radical idea: for one month, stop being the curator of the systems museum and see who you are without the identity of “person with the perfect OS.” for one month, stop being the curator of the systems museum and see who you are without the identity of “person with the perfect OS.” a minimal stack isn’t your final setup for life. it’s rehab after the museum — the first moment you can actually write an honest self.md ruleset minimal stack self.md: a protocol for how you treat yourself self.md self.md is basically my personal README. a living markdown file i can reopen anytime — especially when i’m about to “just add one more project” and pretend it’s self-care. self.md a typical manifesto sounds like: “i’ll wake up at 5am” “i’ll read a book a week” “i’ll run three big projects” “i’ll finally fix health, money, relationships, and career at the same time” “i’ll wake up at 5am” “i’ll read a book a week” “i’ll run three big projects” “i’ll finally fix health, money, relationships, and career at the same time” these texts go straight to the museum: a beautiful Notion page you’re ashamed to open in a month. self.md is a different form. self.md it’s not “how i become ideal.” it’s: what i consider acceptable vs unacceptable treatment of myself for the next few years. it includes rules like: it includes rules like: “i don’t agree to work setups where my calendar is 100% meetings.” “after 22:00 on weekdays, i don’t make strategic decisions — i only capture thoughts.” “i can’t have more than three active growth directions at once.” “i use AI for unloading and honest math, not for generating another ideal life.” “i don’t agree to work setups where my calendar is 100% meetings.” “after 22:00 on weekdays, i don’t make strategic decisions — i only capture thoughts.” “i can’t have more than three active growth directions at once.” “i use AI for unloading and honest math, not for generating another ideal life.” one Space participant wrote after the first article: one Space participant wrote after the first article: “the idea that a person should have a system prompt too is very close to me. i literally have that text and try to reread and update it — special notes in Obsidian.” — Kirill Oleinichenko “the idea that a person should have a system prompt too is very close to me. i literally have that text and try to reread and update it — special notes in Obsidian.” — Kirill Oleinichenko so for some people, a system prompt already exists as a working document — not a pretty manifesto. if you don’t have one yet, write it not as “ideal me,” but as your internal board of directors. self.md as a board meeting for many people, the roles look like this: cynical Manson who cuts deep: “this is ego” the tired one who just wants sleep and to not die rational product manager who looks at metrics anxious one who’s afraid of losing money and status cynical Manson who cuts deep: “this is ego” the tired one who just wants sleep and to not die rational product manager who looks at metrics anxious one who’s afraid of losing money and status practice: self.md 2026, board protocol practice: self.md 2026, board protocol list 3–5 internal characters (you can borrow real authors/mentors as masks) give each one 1–2 rules for 2026, for example: list 3–5 internal characters (you can borrow real authors/mentors as masks) give each one 1–2 rules for 2026, for example: manson: “no projects for status only. if i can’t explain why it’s mine — i don’t take it.” tired one: “no more than N evenings of work per week. if i break it — it’s a system bug, not a moral failure.” product: “if my calendar has no focus blocks, only meetings — that’s firefighting mode, not normal.” anxious: “i don’t promise anything until i’ve done the time math. first calculation, then ‘yes.’” manson: “no projects for status only. if i can’t explain why it’s mine — i don’t take it.” manson: tired one: “no more than N evenings of work per week. if i break it — it’s a system bug, not a moral failure.” tired one: product: “if my calendar has no focus blocks, only meetings — that’s firefighting mode, not normal.” product: anxious: “i don’t promise anything until i’ve done the time math. first calculation, then ‘yes.’” anxious: assemble 7–10 points into your self.md. not pretty — operational. anything that smells like “on monday i’ll become a perfect human” goes in the trash (or straight into the museum).anything that sounds like real constraints and real agreements goes into your prompt. assemble 7–10 points into your self.md. not pretty — operational. self.md anything that smells like “on monday i’ll become a perfect human” goes in the trash (or straight into the museum).anything that sounds like real constraints and real agreements goes into your prompt. AI: museum painter or honest x-ray? in this story, AI plays two roles. the unhealthy role: museum painter. the unhealthy role: museum painter. that’s when you ask: “build me the perfect Notion system” “design a schedule for a hyper-productive person” “give me 50 habits to become a top manager” “build me the perfect Notion system” “design a schedule for a hyper-productive person” “give me 50 habits to become a top manager” if you ask AI to design an ideal life before you’ve chosen your projects and constraints, you’re not solving the problem — you’re automating your context narcissism. the healthy role: x-ray + secretary. the healthy role: x-ray + secretary. that’s when you say: “here’s my task list and hours. calculate the real total.” “here are my projects. help me sort them into mine / imposed / phantom.” “here’s my self.md. remind me when i’m trying to stuff in a fourth project.” “here’s my task list and hours. calculate the real total.” “here are my projects. help me sort them into mine / imposed / phantom.” “here’s my self.md. remind me when i’m trying to stuff in a fourth project.” self.md in that role, AI: reveals absurdity (hello, 29.5-hour day) helps unload context and shape it into something usable returns you to your own rules instead of designing a new ideal museum reveals absurdity (hello, 29.5-hour day) helps unload context and shape it into something usable returns you to your own rules instead of designing a new ideal museum if you want a simple “what to delegate vs keep human” rubric (so AI doesn’t become your museum interior designer), this is the cleanest cheat sheet i’ve seen: https://self.md/concepts/delegation-principles/ if you want a simple “what to delegate vs keep human” rubric (so AI doesn’t become your museum interior designer), this is the cleanest cheat sheet i’ve seen: https://self.md/concepts/delegation-principles/ https://self.md/concepts/delegation-principles/ where {context} lab fits, and what’s next this essay came out of the AI Mindset ecosystem — specifically, watching smart people accidentally build beautiful cages for themselves. at some point we even ran a time-boxed sprint called {context} lab to help people shut down their productivity museums and rebuild a minimal stack. that cohort is closed now, but the move still works: beautiful cages if you combine both articles, the picture is pretty honest: if you combine both articles, the picture is pretty honest: you’re not burned out — you’re living with context obesity a big chunk of that obesity is your systems museum, fueled by context narcissism the exit isn’t “another productivity course,” it’s: a minimal stack you can actually carry an honest self.md as a protocol for treating yourself AI as an assistant-x-ray, not a museum painter you’re not burned out — you’re living with context obesity a big chunk of that obesity is your systems museum, fueled by context narcissism the exit isn’t “another productivity course,” it’s: a minimal stack you can actually carry an honest self.md as a protocol for treating yourself AI as an assistant-x-ray, not a museum painter a minimal stack you can actually carry an honest self.md as a protocol for treating yourself AI as an assistant-x-ray, not a museum painter a minimal stack you can actually carry an honest self.md as a protocol for treating yourself self.md AI as an assistant-x-ray, not a museum painter in this picture, {context} lab was a space where participants temporarily close the museum entrance, make a few painful decisions in a safe environment, assemble their stack and their self.md — and only then attach AI assistants and automations. {context} lab self.md and for you, the choice is weirdly binary: you can close this tab and do a small Notion cleanup or today — honestly turn off at least one system for a month, write down your three projects, and draft the first five lines of your self.md self.md in the second scenario, a year from now you’ll have a much better chance of living your life — not someone else’s expectations. Ray Svitlastay evolving 🐌
