The Ideal Lawn Mower Doesn’t Exist
What a 40-year-old TRIZ argument about grass teaches you about every engineering debate you’ll ever have
There’s an old argument inside TRIZ that sounds like a joke the first time you hear it, and then quietly stops being one.
The question: what’s the ideal solution to cutting grass?
Not the best mower. Not the quietest, fastest, most autonomous mower. The ideal solution has no mower at all. The grass simply stops growing at two inches, the way a human stops growing at some final height, and the function — a manicured lawn — exists without anyone or anything performing an action to produce it.
The opposition’s response is immediate and reasonable: that’s an enormous biomimetic investment in genetic engineering just to avoid buying a mower.
Both sides are right. That’s what makes the argument worth having.
What the process believed
Every mower ever built — push, self-propelled, riding, robotic — has been an answer to the same unexamined question: how do we cut grass better? Faster, quieter, less fuel, less labor, more autonomy. Each generation’s “delighter” becomes the next generation’s assumed minimum — self-propulsion was a marvel once, then a baseline. The robotic mower quietly running laps while you’re at work is this decade’s version of the same story, and in ten years it’ll just be what a lawn mower does.
Every one of those improvements optimizes the mechanism. None of them asks whether the mechanism should exist.
That’s the question TRIZ’s Ideal Final Result forces onto the table: not “how do we build a better mower,” but “what if the function existed and the mechanism didn’t?” IFR doesn’t want a better tool. It wants the problem to solve itself, or better yet, to stop needing solving.
The contradiction
Here’s the pair actually in tension: elegance of the ideal vs. cost of getting there.
Self-limiting grass is IFR taken to its logical end — zero mower, zero fuel, zero noise, zero labor, function still delivered. On paper, that’s the most elegant answer in the room. The objection — too much biomimetic investment for the payoff — isn’t wrong, but it’s usually stated as a feasibility complaint when it’s actually an objective-function disagreement in disguise. It’s really saying: the one-time R&D cost of engineering a stop-condition into grass exceeds the value of the outcome.
That framing hides its own contradiction. It measures the numerator — the R&D cost — against a single mower, when the honest comparison is against every mower, every gallon of gas or battery charge, every hour of human or robotic labor, on every lawn, forever. TRIZ has a name for that kind of unexamined assumption: mistaking a recurring cost for a one-time one, simply because the recurring cost is familiar and the one-time cost is unfamiliar and therefore feels bigger.
But the opposition has a stronger argument available, if it reaches for it — one that isn’t about money at all. Human height plateaus because of a hormonal feedback system that evolved for reasons that have nothing to do with landscaping convenience. Grass has no equivalent internal reason to stop growing at a height humans find tidy. Cutting height is a human preference, not something any selection pressure in the plant’s evolutionary history was ever pushing toward. So the real floor isn’t “is this too expensive” — it’s “does a trait like this exist anywhere in nature to borrow from at all.” For self-limiting height in turfgrass, specifically, the honest answer is close to no.
The lessons
1. Every “ideal final result” argument eventually reduces to a resource question — but not always the resource you think. The first objection reaches for cost. The stronger objection reaches for biological precedent: is there a selection pressure anywhere in nature this trait could attach to, or are you asking evolution to install a stop-condition it never had a reason to build.
2. Comparing a one-time investment to a recurring cost without amortizing is a quiet accounting error, and it shows up everywhere, not just in lawn care. Any time “too expensive” is the whole objection, ask: expensive compared to what, over what time horizon, against how many repetitions of the current cost.
3. Delighters expire on a schedule, and the schedule is the whole story. Self-propulsion, then autonomy, then — maybe — the mechanism disappearing altogether. Each stage makes the previous one look primitive in hindsight. The lawn mower’s history isn’t a sequence of better tools. It’s a sequence of the tool mattering less.
4. The “ideal” isn’t a destination, it’s a direction-finder. Nobody is engineering self-limiting grass this year. But asking the question anyway reorganizes every intermediate step — a mower that needs less human attention is a step toward the ideal, not just a better mower, once you’ve stated where “better” is ultimately headed.
The principle
The Best Answer to “How Do We Do This Better” Is Sometimes “What If We Didn’t Have To.”
Every engineered product carries an assumption about what its mechanism is for — and the assumption is rarely questioned, because the mechanism has always been there. TRIZ’s discipline is asking the question anyway, all the way out to the edge of absurd, because the answer at the edge tells you which direction “better” has been pointing the whole time.
Before you go
Pick something in your life you improve constantly without ever asking whether it needs to exist at all — a routine, a tool, a recurring task you’ve gotten steadily better at doing.
What would the ideal version look like if the task simply stopped needing to happen?
And be honest: is the obstacle in the way of that ideal a cost problem — or is it something more fundamental, like nothing in the system was ever built to want that outcome in the first place?
A special note:
A thank-you: Ellen Domb was one of my first TRIZ teachers, and she was the one who first posed the self-limiting grass question to me, years ago. She passed away last December. Every time this argument comes up, it's hers before it's mine — I'm just still turning it over the way she taught me to.
Herbert Roberts, P.E. — 30+ years in aviation R&D across two companies. 8+ years analyzing accidents for attorneys under my P.E. license.