Working on a project of your own — "skating" rather than "walking" — is not just more fun but dramatically more productive. Yet school, social custom, and adult perfectionism systematically suppress this mode of work.
Paul Graham argues that a specific kind of autonomous, self-directed work — a "project of one's own" — produces the most important achievements and the deepest engagement, yet it is actively discouraged by educational systems and social norms that treat hobbies and play as separate from "real work." The essay traces the nature of this work, how organisations can preserve it at scale, and how adults can recover the careless confidence of childhood treehouse-building to start projects that matter.
Skating vs. Walking — a qualitative difference, not a matter of degree. Graham introduces a central metaphor: working on a project of your own is as different from ordinary work as skating is from walking. Skaters are not necessarily happier — they are often worried and puzzled during the process — but they feel "awake and alive," doing what they were meant to do.[1] This state cannot be produced by external incentives; it emerges from ownership and voluntary engagement.
Two axes of "one's own": voluntary vs. solo. Graham identifies two distinct senses: (a) doing work voluntarily rather than because someone told you to, and (b) doing it by yourself rather than as part of a collective. The first boundary is sharp — people sensitive to autonomy can instantly feel the difference between pulling and being pushed. The second is soft, shading gradually into collaboration via shared projects or complementary puzzle-piece work.[2] The critical variable is autonomy, not isolation.
School is the enemy of skating — and this is not accidental. Schoolwork is typically neither a project nor one's own. Graham observes that treating "playing" and "hobbies" as qualitatively different from "work" conceals the direct route from a childhood treehouse to architecture or engineering. The path we point kids to goes through grades and exams instead, even though "the work that made Darwin and Newton famous was actually closer in spirit to building treehouses than studying for exams."[3] Graham would rather his kids build ambitious projects than get good grades — and reports that at Y Combinator, he cared only about applicants' projects, not their grades.
The Macintosh team preserved skating at scale. The original Macintosh team — Burrell Smith, Andy Hertzfeld, Bill Atkinson, Susan Kare — exemplified how organisations can maintain the "project of one's own" feeling. They were not employees following orders but "rockets let loose by Steve Jobs" who came back after dinner to work late into the night out of excitement, not compulsion.[4] Graham uses this to challenge dogmatic "work/life balance": for skaters, work and life are connected by a dash, not a slash.
Clean-sheet design is a feature, not a bug. Programmers' tendency to rewrite from scratch often alarms managers, but Graham argues it's partly driven by the greater reward of skating. Writing code from scratch is so much more rewarding that a good programmer can end up net ahead despite the waste. This natural alignment between skating and solving new problems is one reason startup payoffs are so high: unsolved problems command a market premium, and you get a productivity discount when skaters work on them.[5]
The inception phase is where projects die — and high standards are the killer. Graham identifies the critical moment as "when you go from thinking it might be cool to do x to actually doing x." At this point, high standards are "not merely useless but positively harmful."[6] Adults are deterred by fear of failure from starting projects that would have succeeded if they had. Kids plunge into treehouses without worrying about waste or comparison — and Graham argues adults should deliberately recapture that "careless confidence."
"Working on a project of your own is as different from ordinary work as skating is from walking."
— Paul Graham[1]"For those to whom the word 'work' automatically implies the dutiful plodding kind, they are [distinct]. But for the skaters, the relationship between work and life would be better represented by a dash than a slash."
— Paul Graham[4]"If I had to choose between my kids getting good grades and working on ambitious projects of their own, I'd pick the projects."
— Paul Graham[3]"They were not tennis balls hit by Steve Jobs, but rockets let loose by Steve Jobs."
— Paul Graham, on the original Macintosh team[4]"The most important phase in a project of one's own is at the beginning: when you go from thinking it might be cool to do x to actually doing x. And at that point high standards are not merely useless but positively harmful."
— Paul Graham[6]
✓ VERIFIED — The word "hobby" originally referred to a hobby-horse ridden by a child, and by metaphorical extension an obsession or pursuit. The etymology is consistent with Graham's account: "hobby" came to mean a favourite pursuit via this metaphor.[7]
✓ VERIFIED — Andy Hertzfeld's book Revolution in the Valley (2004) documents the Macintosh team's intense, self-driven work habits, including late-night coding sessions fuelled by excitement rather than obligation. Graham's account is consistent with primary sources.[8]
⚠ UNVERIFIED — The claim that "what proportion of great work has been done by people who were skating in this sense? If not all of it, certainly a lot" is an opinion claim that cannot be empirically verified without defining "great work" and measuring motivational states retrospectively.
For knowledge workers and creatives: Audit your current portfolio of work for the skating/walking ratio. Tasks you resent are walking; tasks you lose track of time in are skating. Shift toward the latter — even if it means declining assignments or changing roles.
For managers and founders: The lesson from Jobs and the Macintosh team is that you can scale skating by recruiting people who already have the orientation, then giving them high-level direction and getting out of the way. Bureaucracy and excessive process are the enemies of this mode.
For parents and educators: Graham's argument is a direct challenge to credential-focused parenting. If you want your child to do great work, protect their project time as seriously as their homework time — perhaps more seriously. And crucially: never force projects, because that destroys the very autonomy that makes them powerful.
For anyone stuck not starting: The single most actionable insight is that high standards are harmful at the inception phase. The barrier to starting is almost always internal perfectionism, not external constraints. Start before you're ready.
Source credibility: High — Paul Graham is a well-known essayist, programmer, and co-founder of Y Combinator. His track record on startup-related insight is established. The essay draws on his direct experience evaluating founders and on primary sources about the Macintosh team.
Claim verifiability: 4 of 5 key claims verified or accepted as authorial opinion. The etymological and historical claims check out against external sources.
Potential biases: The author's perspective is shaped by his experience in startups and tech. The essay implicitly valorises a Silicon Valley / founder-oriented approach to work. There is little engagement with structural constraints that prevent many people from choosing their projects (e.g., economic necessity, caregiving responsibilities). The essay also risks romanticising overwork by framing the dash-between-work-and-life positively.
Quality flags: None. The source is a clean, well-written essay with linked footnotes.
Confidence in synthesis: High — the essay's argument is clear, the structure is explicit, and the key insights are directly stated.
Steelman critique: The essay's celebration of "skating" and the blurring of work/life boundaries is a luxury good. Most people cannot choose their projects — they work to pay bills, often in roles with little autonomy. By framing those who do assigned work as mere "plodders" or "walkers," the essay risks classism and ignores that the institutional structures (grades, credentials, hierarchies) Graham dismisses exist precisely because they provide reliable paths for people without the safety net to skate. Furthermore, some of the world's most important work — maintaining infrastructure, nursing, teaching, sanitation — is deeply un-skate-like and yet indispensable.
What would need to be true: For Graham's vision to generalise beyond the tech/startup world, economic security would need to be decoupled from credential-based employment, social safety nets would need to be strong enough that failure of a self-chosen project is not catastrophic, and organisational structures would need to be redesigned around autonomy rather than control — all of which are systemic changes far beyond individual choice.
[1]: [Paul Graham, "A Project of One's Own"] The skating/walking metaphor and the description of the skater's emotional state.
[2]: [Paul Graham, "A Project of One's Own"] The two axes of "one's own": voluntary vs. solo, with collaboration as a gradient.
[3]: [Paul Graham, "A Project of One's Own"] School's suppression of projects; Darwin/Newton comparison; YC preference for projects over grades.
[4]: [Paul Graham, "A Project of One's Own"] Macintosh team as exemplar; work/life dash rather than slash.
[5]: [Paul Graham, "A Project of One's Own"] Clean-sheet design and the startup productivity double-increase.
[6]: [Paul Graham, "A Project of One's Own"] The inception phase and the harm of high standards.
[7]: [Etymonline, "hobby (n.)"] Etymological confirmation of the hobby-horse origin.
[8]: [Andy Hertzfeld, Revolution in the Valley (O'Reilly, 2004)] Primary-source account of the Macintosh team's self-motivated work habits.
Generated by OmniMiner v7.2 · openai/gpt-oss-120b · 2026-06-14