This is a very different problem than actual efficiency when mowing a lawn or vacuuming a rub. Turning takes more work and time, and you miss a part of the 'square' when you are turning (the arc of the curve is not a straight line). In addition, vacuums especially have an area at the edge of the machine that is not cleaned as well, which is why you want a bit of overlap in your lines.
Plus, when mowing a lawn, you often want to leave a nice pattern, so you cant just go for pure efficiency.
When I was laid off a couple months ago, the first thing I did was go out and mow my lawn. One hour of simple, clean labor, with clear progress, goals, and solitude. It helped me stay out of a bad mental frame, and when I finished I went inside and sent out a few hundred applications
Over the next several weeks, I spent a lot of time that wasn't spent on finding a new job working on my yard. Fixing sprinklers, cleaning up the edges of the grass, turning, peating, and mulching garden beds, and more
While this job search period took longer than any of my past ones, and ostensibly should have been more stressful, it wasn't. Is that because I spent most of the downtime working outdoors? Probably, who knows? It was certainly a better use of my time than sitting indoors moping
Niceness of lawn mowing is very climate dependent. Around 70 degrees I don’t mind it. Different story when it’s 98.
Unenjoyable, and my electric motor’s batteries die twice as fast in the heat, so I have to do it two days in a row instead of being able to get it over with. But I don’t hate it quite enough to blow hundreds of dollars on extra batteries or to pay someone else to do it.
A friend of a friend got sick of tech work. Now she bills out at $70-120/hr doing garden and landscaping work. Not her crew, just her. Solo work on nice houses. Not lawn-mowing, landscaping. Very dirty every day. If I had a job I'd be jealous.
Absolutely agree. I could easily afford to hire a lawn crew to cut my lawn, but I enjoy doing it every weekend or two for similar reasons. I have a little under half an acre to mow, but I still enjoy doing it manually with a push mower.
Gets me outside and moving, accomplishes something tangible, can catch up on a podcast or two or just get lost in my thoughts, often leads to fixing other odds and ends around the house I've put off.
Even when it comes to mowing, different people have different goals. I live out in the country, and tall grass means bugs and ticks, so my mowing tries to strike a balance of short grass near the home and the property edges to stop the ticks, medium grass most everything else because we like to let the clover grow, and tall grass, un-mowed for years on some spots, to allow an ecosystem where we still get monarchs and nightly fireflies.
Totally understandable if suburban folk have different goals, but either way, 100% agreed that efficiency is a secondary goal at best.
This is mentioned in the article when discussing a player that found the optimum route:
> Sarah also did something that appeared a bit unnatural. Instead of simply continuing straight a few times, she intuitively and preemptively turned in a snake-like pattern. This highlighted a key difference between real-life mowing and our simplified game, an approach that strong mowers exploited. Sarah, an avid gamer, seemed to recognize this strategy immediately.
The thing is, one needs to consider more than efficiency, but the overall health and appearance of the lawn --- as noted on _Magnum PI_, it is necessary to vary the cutting so that the lawn does not develop a "grain" where it lays unevenly.
You should see how optimal the poster's first dates are! Meet at a location with premade bottles of Soylent while they ask 15 questions. Done in 30min. Optimized away the cost and time "wasted" at restaurants or mini-golf type activities.
It's a cute little optimization puzzle they made for people to solve, with lawn mowing as a graphical hook to make the problem more approachable and fun. Similarly, my kids have a physical logic puzzle where the theme is putting plastic cats into plastic boxes, but it's not supposed to be training for actually optimizing cat packaging.
How is this any more or less valid a use of time vs. mini-golf?
(The "real" analogous problems are things like optimizing factories or delivery routes, and the way they are solved is by programming a computer to, but this is a toy version.)
I just send prospective dates a pre-screening questionaire, my social résumé, and a doordash giftcard. It's really efficient since I don't need to waste my time going on bad dates or even leaving the house. Unfortunately I haven't figured out the best way screen for candidates on the second round because they never return the questionaire.
That’s rather cruel. Optimizing lawn mowing conserves human and machine fuel/battery, especially on large plots or as professionals or in hot climates. Misapplying optimizations to one’s free time is absolutely in line with your stereotype, but that’s much more a widespread defect of judgment in computer engineers than in mowers.
I was thinking the same thing. When actually mowing a lawn (or similarly, vacuuming a carpet/rug) I tend to optimize for either long lines or rectangular sections where I can easily mow a straight pattern.
> and you miss a part of the 'square' when you are turning (the arc of the curve is not a straight line)
This can heavily be mitigated through technique. If you walk straight and just turn? Yeah, the corner is missed. If you explicitly come to a stop and do your best to do a full 90-degree turn before moving? The only curve there is going to come from the shape of the blade, and is almost guaranteed to be a tighter curve than just walking and turning without stopping.
Yes, it takes longer. But if it's worth doing, it's worth doing right, even if it takes twice as long, no?
> Yes, it takes longer. But if it's worth doing, it's worth doing right, even if it takes twice as long, no?
This is debatable, but either way it is the opposite of the purpose of this article, which is about 'efficiency' in mowing.
Also, even if you do it very carefully, and walk a perfect 90 degree turn, you are still going to have an arc. You wont mow the whole 'square' like in the game.
Semi-related: Worked as a landscaper during my youth, one of my first jobs
I used to think that the line patterns mowed into grass were a purely aesthetic thing.
It turns out, the lines are mowed in a rotation to prevent wearing out the grass and causing damage. First you mow length-wise, then width-wise, then diagonally, finally diagonally in the opposite direction, before cycling through again.
Absolutely correct. If you always mow in the same direction, the grass always gets pushed in the same way by the lawnmowers blades and it will eventually morph to that direction and lose strength, which appears as grass that doesn't stand straight up, and it looks worse.
I don't usually do the diagonal paths but I do try to go at right angles to the previous mowing each time I mow the lawn (excepting when I can't remember which way I did it last time, but it's usually still visible) or a few narrow areas which really only make sense to mow in one direction.
You worked for someone better than average, or I worked for someone worse than average, because this sounds like a thoughtful touch and they never told us anything like that.
I have a battery electric lawnmower. I found I got noticably better battery life after sharpening the blade to a near razor edge. Yes the edge is more fragile but it cuts the grass with much less resistance.
I believe the point is the wheel lines and to vary where the wheels go by changing the pattern each time you mow so you don't, for example, always drive exactly down the fence and cause a rut or mulch buildup or missed area.
I dont think it was "don't hurt the grass while you cut it" or "mow it four times in one sitting".
We people vary frequency with weather. Lawn tractors are heavy and, due to moisture, ground isn't always hard. On large properties some of the ground stays moist longer due to drainage issues. So yes, after years of indifferent mowing, one can detect where wheels have been moving topsoil.
They don't mean do all those cuts each time you mow. You mow the grass in one direction one day, and then a few days later when it needs cut again, you mow a different pattern.
When I used to mow neighbor's yards as a kid, I'd do a crosshatch pattern that I'd rotate 45deg each week, orthagonal vs diagonal. Just doing one direction looks fine but a crosshatch looks even better, especially if they have thick, healthy grass. I wouldn't be as meticulous on the side and backyards, those would just get a single, slightly overlapping pass in whatever direction was convenient unless they had a really nice backyard, then I'd do extra there.
People like that golf course look and I'd get nice tips for doing it.
I do this very thing in the fall when all of my neighbor's trees drop their leaves which blow into my yard. I don't bag leaves. I mulch them. To get them down to a respectable looking yard, it can easily take this many passes. I do it all over again in the early spring towards the end of February when the Live Oak trees in my yard finally drop their leaves for the year. According to my watch, I can cover more than 2 miles just taking my mower for a walk on those days.
I started doing it and immediately felt like I was optimizing for minimal moves like a video game, rather than what I'd actually do which is not the same strategy. There's a handful of things different
- The direction you mow leaves a pattern that I generally try to stay parallel to, and mow alongside, and go for long strips even if I have to do a bit of running over already-mowed parts
- I may go diagonal at times
- I've never had a yard that's a maze of rocks
- If there's a small patch I need to mow around, like a garden, I'll likely do a circle around that and use the edge of that mow line as a new edge
I'd bet the vast majority of people who did the game input a series of moves that doesn't reflect that closely to what they'd do in real life.
It wasn't even clear how they were going to score it, or if it was said then I didn't see it.
I kinda guessed they would score it the way they did - the number of squares you traveled through, but at the same time I was half aware of doing what I would do in real life which is to minimize the number of direction changes, not distance traveled.
WJW. This entire article is assuming there’s a single measure of “success” in this endeavor and the simulation isn’t remotely anything like mowing an actual lawn.
I’m usually a huge fan of Pudding, but this completely misses the mark.
I have a small yard and I'm not optimizing for the fewest moves, I'm optimizing for both the longest continuous lines (because I think they look nicest) and the shortest distance to carry the clippings at the end.
I have a different definition of optimal to this game.
Yeah, I think most people who have a yard wouldn't have played it for fewest moves... I don't know why but this whole piece reminds me of someone that never had to mow their yard, or AI that optimizes for something without carrying the full context or nuance of why we might not have gone with fewest moves optimization.
I'm right with you. My dad was a farmer and taught me: First deal with the irregular, annoying parts on the outside. Create rectangles with a long side and turning space on each shorter side. Then deal with each rectangle by going back and forth along the long edge. The turnarounds are also a great place for some wheelbarrows to collect the clippings closest to the Komposthaufen.
Hence why I went for the irregular strange part on the lower left first, and then the more regular part on the right. That's the weird rain collection, herb garden, thing we have at the back.
> Then deal with each rectangle by going back and forth along the long edge.
I prefer to work in lands after I have cut the headlands. Reduces turning stress on the grass and acknowledges that the machine (the one I own, but the design is common) is optimized for use in one direction.
I always mow a different path each time because turns and wheels slowly build up damage to the lawn and so I want to spread that out over everything. Push mowers are light enough that this doesn't matter but a rider it does.
Conversely, I once had a very large yard for a suburban home, about a third of an acre. I sectioned my yard into about 5 sections, but I mowed pretty much the same way: whatever lines are longest without stopping.
The less time I spent in Alabama heat, the better.
This is a really nice presentation of how some people play the authors' lawn-moving game better than others -- a great analysis of the gameplay. But the majority of comments on this HN thread right now are reacting to the title rather than to the article, as if it were about real-life lawn-moving.
And not enough people are reacting to the rocks that should just be removed in the first place to make the optimised algorithm a lot simpler. And I thought I was on HN.
There's a spot where you could just lift up the mower over the rocks to get from one side to the other. Or at least, that's how I would have done, but I have a tiny lawn and a reel mower.
There are a number of bad premises in this article.
For example, most people in the world (outside the USA in particular) have never mowed a lawn. I live in a city of 3.5 million and essentially no one has a lawn.
When you really mow a lawn terrain matters. Most lawns are not flat. One goes on lines of equal height not up and down a hill.
Edges are different than the interior part of the lawn, particularly if they abut flower beds, curbs, etc.
Re passing a given spot is sometimes an efficient way to mow it properly. Particularly in dead ends and odd corners.
Etc.
In addition to the obvious things already mentioned, the authors miss a subtle but important point. Regardless of one’s preferences for efficiency, turning, patterns, grooves, etc.. once you are done mowing the lawn you don’t just leave the mower in the yard and go to sleep. Their ‘perfect’ solutions take 11 steps longer than finishing on the right.
Reminds me of Zachtronics: Take a computationally difficult problem, reduce it to a game, and then score people against each other.
Pure genius and nerd snipe marksmanship.
Did you know multi-robot exploration pretty much as laid out in Exapunks has loose but well studied bounds on number of agents, bits of information, and exploration steps?
For example, one agent with only one pebble cannot explore all graphs?
This is exactly the trick one of my colleagues uses to teach Theory of Computation, a notoriously boring subject (for your average CS student). Instead of motivating reductions by blathering on about how knapsack and TSP are the same, let them choose a fun video game and show how it can be solved using integer programming or something. Students have way more fun and end up actually enjoying the proofy parts, because they have actual practical value.
Though I solved it in 49 moves (yay me), I realized when thinking about efficiency, it didn't take into consideration two important things:
1) all of the turns I was making in order to be most efficient, whereas overlapping a few times would have probably taken less time, and
2) I spent a lot of thinking on the spot, as to what was the most efficient path, rather than moving the mower.. so while I only took 49 moves, the time spent on inaction may have balanced the scales with someone who just kept moving.
There's a life lesson in there somewhere, I'm sure!
> Sarah, an avid gamer, seemed to recognize this strategy immediately.
As soon as I read this, I immediately thought of the puzzle in the last gym of the Gen 3 Pokémon games (Ruby/Sapphire/Emerald). It was a big ice platform that cracked after stepping on it once, and backtracking meant falling through the ice and having to start all over again.
It made me think of the “step on all the colored tiles without crossing your path” puzzles in Links Awakening!
I wonder how the optimal path changes if turns are penalized as well as steps - in real life turning a lawnmower takes a lot more effort than moving the same spot twice in a straight line.
A lawn mower cuts grass whether you push it forward or pull it backwards: my trick when I was a kid was just to push and pull it back and forth across the lawn without changing the direction it faced at all.
You mean a flymo. It's more hovercraft than drone, and much older technology. Works reasonably well on non-lumpy lawns - my neighbor had one in the 1980s.
have to say i really enjoyed how this was presented.
no comment on the results or the premise of are people good at mowing or just good at playing "snake"
i would be very curious to see if this same result held up when you also factor things like turn effort or "prioritize the longest straight lines bc they look cool"
I'm sure I'm the odd ball, but ever since I was a kid, I liked the diagonal patterns much more than the square lines.. and I'm not just talking about lawns. I would vacuum my room on diagonals.. lol
Mowing the lawn is never about mowing but people wanted to be busy, feel productive for a bit, some exercise, and helping to clear thoughts. For others it might be driving long roads, or doing chores inside the house, cycling, and such. Otherwise, it’s inefficient and useless, the lawn can be replaced with better flowers or even small garden, or simply pave it and save the time and efforts and electricity/water doing a repetitive thing like it, and also spare your neighbors the noise 7am Saturday morning, and if you look at the bigger picture, the process is a total waste, you waste water to grow crops only to waste electricity to mow it later, it provides no useful product or outcome either, no flowers, no food, not even cosmetics like fancy looking garden, plus the maintenance and/or initial installations.
Probably the same reason some food delivery people place my bag next to the door instead of in front of the outward opening door where it will tip over, unsure what that reason is but I’d like to know.
> You’ve probably done something like mow a lawn or vacuum a rug hundreds of times without thinking much about it. Some part of your brain works out a route that’s usually good enough to get the job done. People are actually pretty good at this, and better than you’d guess for a problem that is hard for computers. We wanted to see it for ourselves, so we built a lawn and asked people to mow it.
Is there supposed to be more content than this?
... If JavaScript is required to view more content than this, why?
> ... If JavaScript is required to view more content than this, why?
By "asked people to mow it" they're talking about the visitors to the site. It's interactive.
It's somewhat a mistake not to have "skip to results" in the actual html. On the other hand, it makes sense they want to force people to see the landing page first. And unless they want to add a whole rendered fallback the results page isn't going to be super functional without javascript.
Without JavaScript, there is literally no page text beyond that point until the footer. Thee is no indication that there would be any interactive content. There is also no discussion of what the interactive content demonstrates after it. There are no "results" to which a "skip to results" could skip.
Yeah I know. But "literally no page text" is only missing a couple words. The correct version is super short.
The noscript version needs a button and a big box with an X. And even then the button would only be half-functional so I understand skipping it. It mostly just needs a "missing content" warning.
This is a very different problem than actual efficiency when mowing a lawn or vacuuming a rub. Turning takes more work and time, and you miss a part of the 'square' when you are turning (the arc of the curve is not a straight line). In addition, vacuums especially have an area at the edge of the machine that is not cleaned as well, which is why you want a bit of overlap in your lines.
Plus, when mowing a lawn, you often want to leave a nice pattern, so you cant just go for pure efficiency.
Plus mowing a lawn is meditative
Copying another comment I made earlier
When I was laid off a couple months ago, the first thing I did was go out and mow my lawn. One hour of simple, clean labor, with clear progress, goals, and solitude. It helped me stay out of a bad mental frame, and when I finished I went inside and sent out a few hundred applications
Over the next several weeks, I spent a lot of time that wasn't spent on finding a new job working on my yard. Fixing sprinklers, cleaning up the edges of the grass, turning, peating, and mulching garden beds, and more
While this job search period took longer than any of my past ones, and ostensibly should have been more stressful, it wasn't. Is that because I spent most of the downtime working outdoors? Probably, who knows? It was certainly a better use of my time than sitting indoors moping
I always thought this song was about mowing the lawn during tough times, but apparently not:
https://genius.com/The-all-american-rejects-move-along-lyric...
I'm not much of a workaholic, I don't really like going outside or getting my hands dirty...
But mowing the lawn is nice. It's like doing a giant coloring page with a big thick marker.
The past couple of years I've lived in a place with no lawn to mow and I miss it.
Niceness of lawn mowing is very climate dependent. Around 70 degrees I don’t mind it. Different story when it’s 98.
Unenjoyable, and my electric motor’s batteries die twice as fast in the heat, so I have to do it two days in a row instead of being able to get it over with. But I don’t hate it quite enough to blow hundreds of dollars on extra batteries or to pay someone else to do it.
There is something beautiful about getting hands dirty with actual dirt/earth. Its such a REAL feeling.
A friend of a friend got sick of tech work. Now she bills out at $70-120/hr doing garden and landscaping work. Not her crew, just her. Solo work on nice houses. Not lawn-mowing, landscaping. Very dirty every day. If I had a job I'd be jealous.
I bought a real stinker of place to renovate & of course the yard is completely overgrown.
Every wo/man and their dog is telling me to knock it down, have a bobcat to clear the yard and start fresh.
I enjoy the work of clearing a yard by hand, and slowly cleaning the place up.
Absolutely agree. I could easily afford to hire a lawn crew to cut my lawn, but I enjoy doing it every weekend or two for similar reasons. I have a little under half an acre to mow, but I still enjoy doing it manually with a push mower.
Gets me outside and moving, accomplishes something tangible, can catch up on a podcast or two or just get lost in my thoughts, often leads to fixing other odds and ends around the house I've put off.
I would agree with you if I wasn't so allergic to grass.
> Plus mowing a lawn is meditative
Not when its >100F and >30%RH and a bright sun in the sky.
That’s what evenings are for…
> you often want to leave a nice pattern
Even when it comes to mowing, different people have different goals. I live out in the country, and tall grass means bugs and ticks, so my mowing tries to strike a balance of short grass near the home and the property edges to stop the ticks, medium grass most everything else because we like to let the clover grow, and tall grass, un-mowed for years on some spots, to allow an ecosystem where we still get monarchs and nightly fireflies.
Totally understandable if suburban folk have different goals, but either way, 100% agreed that efficiency is a secondary goal at best.
This is mentioned in the article when discussing a player that found the optimum route:
> Sarah also did something that appeared a bit unnatural. Instead of simply continuing straight a few times, she intuitively and preemptively turned in a snake-like pattern. This highlighted a key difference between real-life mowing and our simplified game, an approach that strong mowers exploited. Sarah, an avid gamer, seemed to recognize this strategy immediately.
The thing is, one needs to consider more than efficiency, but the overall health and appearance of the lawn --- as noted on _Magnum PI_, it is necessary to vary the cutting so that the lawn does not develop a "grain" where it lays unevenly.
> Sarah, an avid gamer, seemed to recognize this strategy immediately.
I have also played a lot of ChuChu Rocket!
You should see how optimal the poster's first dates are! Meet at a location with premade bottles of Soylent while they ask 15 questions. Done in 30min. Optimized away the cost and time "wasted" at restaurants or mini-golf type activities.
It's a cute little optimization puzzle they made for people to solve, with lawn mowing as a graphical hook to make the problem more approachable and fun. Similarly, my kids have a physical logic puzzle where the theme is putting plastic cats into plastic boxes, but it's not supposed to be training for actually optimizing cat packaging.
How is this any more or less valid a use of time vs. mini-golf?
(The "real" analogous problems are things like optimizing factories or delivery routes, and the way they are solved is by programming a computer to, but this is a toy version.)
I just send prospective dates a pre-screening questionaire, my social résumé, and a doordash giftcard. It's really efficient since I don't need to waste my time going on bad dates or even leaving the house. Unfortunately I haven't figured out the best way screen for candidates on the second round because they never return the questionaire.
That’s rather cruel. Optimizing lawn mowing conserves human and machine fuel/battery, especially on large plots or as professionals or in hot climates. Misapplying optimizations to one’s free time is absolutely in line with your stereotype, but that’s much more a widespread defect of judgment in computer engineers than in mowers.
I was thinking the same thing. When actually mowing a lawn (or similarly, vacuuming a carpet/rug) I tend to optimize for either long lines or rectangular sections where I can easily mow a straight pattern.
Why not mow in a spiral then hit the 4 corners last?
The lines change mowing so much!
> and you miss a part of the 'square' when you are turning (the arc of the curve is not a straight line)
This can heavily be mitigated through technique. If you walk straight and just turn? Yeah, the corner is missed. If you explicitly come to a stop and do your best to do a full 90-degree turn before moving? The only curve there is going to come from the shape of the blade, and is almost guaranteed to be a tighter curve than just walking and turning without stopping.
Yes, it takes longer. But if it's worth doing, it's worth doing right, even if it takes twice as long, no?
> Yes, it takes longer. But if it's worth doing, it's worth doing right, even if it takes twice as long, no?
This is debatable, but either way it is the opposite of the purpose of this article, which is about 'efficiency' in mowing.
Also, even if you do it very carefully, and walk a perfect 90 degree turn, you are still going to have an arc. You wont mow the whole 'square' like in the game.
I do the outside/borders and around the trees first. Then the turning is over the already mowed area.
Semi-related: Worked as a landscaper during my youth, one of my first jobs
I used to think that the line patterns mowed into grass were a purely aesthetic thing.
It turns out, the lines are mowed in a rotation to prevent wearing out the grass and causing damage. First you mow length-wise, then width-wise, then diagonally, finally diagonally in the opposite direction, before cycling through again.
Useless information for you today.
Absolutely correct. If you always mow in the same direction, the grass always gets pushed in the same way by the lawnmowers blades and it will eventually morph to that direction and lose strength, which appears as grass that doesn't stand straight up, and it looks worse.
I don't usually do the diagonal paths but I do try to go at right angles to the previous mowing each time I mow the lawn (excepting when I can't remember which way I did it last time, but it's usually still visible) or a few narrow areas which really only make sense to mow in one direction.
You worked for someone better than average, or I worked for someone worse than average, because this sounds like a thoughtful touch and they never told us anything like that.
wat. Causing damage to the grass you're hitting with a (relatively) blunt object in the first place? All you did was waste three times the gas.
> hitting with a (relatively) blunt object in the first place
Well there's your problem right there.
Lawn mower blades are meant to be somewhat blunt. Like a butter knife. Too sharp and the blade will chip away on the harder materials.
I have a battery electric lawnmower. I found I got noticably better battery life after sharpening the blade to a near razor edge. Yes the edge is more fragile but it cuts the grass with much less resistance.
That I did not know. I still wouldn't stick a finger near one of course.
I believe the point is the wheel lines and to vary where the wheels go by changing the pattern each time you mow so you don't, for example, always drive exactly down the fence and cause a rut or mulch buildup or missed area.
I dont think it was "don't hurt the grass while you cut it" or "mow it four times in one sitting".
How often are you people mowing your lawn that that's even a concern? It's like this is "surburbs: the problem: the comment section"
We people vary frequency with weather. Lawn tractors are heavy and, due to moisture, ground isn't always hard. On large properties some of the ground stays moist longer due to drainage issues. So yes, after years of indifferent mowing, one can detect where wheels have been moving topsoil.
"Worked as a landscaper during my youth"
>All you did was waste three times the gas.
They don't mean do all those cuts each time you mow. You mow the grass in one direction one day, and then a few days later when it needs cut again, you mow a different pattern.
When I used to mow neighbor's yards as a kid, I'd do a crosshatch pattern that I'd rotate 45deg each week, orthagonal vs diagonal. Just doing one direction looks fine but a crosshatch looks even better, especially if they have thick, healthy grass. I wouldn't be as meticulous on the side and backyards, those would just get a single, slightly overlapping pass in whatever direction was convenient unless they had a really nice backyard, then I'd do extra there.
People like that golf course look and I'd get nice tips for doing it.
I do this very thing in the fall when all of my neighbor's trees drop their leaves which blow into my yard. I don't bag leaves. I mulch them. To get them down to a respectable looking yard, it can easily take this many passes. I do it all over again in the early spring towards the end of February when the Live Oak trees in my yard finally drop their leaves for the year. According to my watch, I can cover more than 2 miles just taking my mower for a walk on those days.
I started doing it and immediately felt like I was optimizing for minimal moves like a video game, rather than what I'd actually do which is not the same strategy. There's a handful of things different
- The direction you mow leaves a pattern that I generally try to stay parallel to, and mow alongside, and go for long strips even if I have to do a bit of running over already-mowed parts
- I may go diagonal at times
- I've never had a yard that's a maze of rocks
- If there's a small patch I need to mow around, like a garden, I'll likely do a circle around that and use the edge of that mow line as a new edge
I'd bet the vast majority of people who did the game input a series of moves that doesn't reflect that closely to what they'd do in real life.
Yeah not realistic until:
* You need to pick up dog poo, dog toys, kids toys and clothes pegs
* You need to move the trampoline and mow that part and put it back.
* Some of the grass is so thick you need to whipersnap it first.
It wasn't even clear how they were going to score it, or if it was said then I didn't see it.
I kinda guessed they would score it the way they did - the number of squares you traveled through, but at the same time I was half aware of doing what I would do in real life which is to minimize the number of direction changes, not distance traveled.
WJW. This entire article is assuming there’s a single measure of “success” in this endeavor and the simulation isn’t remotely anything like mowing an actual lawn.
I’m usually a huge fan of Pudding, but this completely misses the mark.
I have a small yard and I'm not optimizing for the fewest moves, I'm optimizing for both the longest continuous lines (because I think they look nicest) and the shortest distance to carry the clippings at the end.
I have a different definition of optimal to this game.
Yeah, I think most people who have a yard wouldn't have played it for fewest moves... I don't know why but this whole piece reminds me of someone that never had to mow their yard, or AI that optimizes for something without carrying the full context or nuance of why we might not have gone with fewest moves optimization.
I'm right with you. My dad was a farmer and taught me: First deal with the irregular, annoying parts on the outside. Create rectangles with a long side and turning space on each shorter side. Then deal with each rectangle by going back and forth along the long edge. The turnarounds are also a great place for some wheelbarrows to collect the clippings closest to the Komposthaufen.
Hence why I went for the irregular strange part on the lower left first, and then the more regular part on the right. That's the weird rain collection, herb garden, thing we have at the back.
> Then deal with each rectangle by going back and forth along the long edge.
I prefer to work in lands after I have cut the headlands. Reduces turning stress on the grass and acknowledges that the machine (the one I own, but the design is common) is optimized for use in one direction.
Or be like me and just mulch the clippings
(yes it looks uglier that way - and yes I think of mowing the lawn as a chore)
I always mow a different path each time because turns and wheels slowly build up damage to the lawn and so I want to spread that out over everything. Push mowers are light enough that this doesn't matter but a rider it does.
Conversely, I once had a very large yard for a suburban home, about a third of an acre. I sectioned my yard into about 5 sections, but I mowed pretty much the same way: whatever lines are longest without stopping.
The less time I spent in Alabama heat, the better.
This is a really nice presentation of how some people play the authors' lawn-moving game better than others -- a great analysis of the gameplay. But the majority of comments on this HN thread right now are reacting to the title rather than to the article, as if it were about real-life lawn-moving.
And not enough people are reacting to the rocks that should just be removed in the first place to make the optimised algorithm a lot simpler. And I thought I was on HN.
There's a spot where you could just lift up the mower over the rocks to get from one side to the other. Or at least, that's how I would have done, but I have a tiny lawn and a reel mower.
I used to do it as a side wheelie for small enough rocks. Good enough, and made me feel like a pro mower when I was younger.
There are a number of bad premises in this article.
For example, most people in the world (outside the USA in particular) have never mowed a lawn. I live in a city of 3.5 million and essentially no one has a lawn.
When you really mow a lawn terrain matters. Most lawns are not flat. One goes on lines of equal height not up and down a hill.
Edges are different than the interior part of the lawn, particularly if they abut flower beds, curbs, etc.
Re passing a given spot is sometimes an efficient way to mow it properly. Particularly in dead ends and odd corners. Etc.
Was hoping this was going to be about a real lawn!
Surprising how good the average person is. I got the median 91% eff. Or 5 off.
In addition to the obvious things already mentioned, the authors miss a subtle but important point. Regardless of one’s preferences for efficiency, turning, patterns, grooves, etc.. once you are done mowing the lawn you don’t just leave the mower in the yard and go to sleep. Their ‘perfect’ solutions take 11 steps longer than finishing on the right.
Reminds me of Zachtronics: Take a computationally difficult problem, reduce it to a game, and then score people against each other.
Pure genius and nerd snipe marksmanship.
Did you know multi-robot exploration pretty much as laid out in Exapunks has loose but well studied bounds on number of agents, bits of information, and exploration steps?
For example, one agent with only one pebble cannot explore all graphs?
https://josh.vanderhook.info/media/pdf/exapunks-graph.pdf
This is exactly the trick one of my colleagues uses to teach Theory of Computation, a notoriously boring subject (for your average CS student). Instead of motivating reductions by blathering on about how knapsack and TSP are the same, let them choose a fun video game and show how it can be solved using integer programming or something. Students have way more fun and end up actually enjoying the proofy parts, because they have actual practical value.
Though I solved it in 49 moves (yay me), I realized when thinking about efficiency, it didn't take into consideration two important things: 1) all of the turns I was making in order to be most efficient, whereas overlapping a few times would have probably taken less time, and 2) I spent a lot of thinking on the spot, as to what was the most efficient path, rather than moving the mower.. so while I only took 49 moves, the time spent on inaction may have balanced the scales with someone who just kept moving.
There's a life lesson in there somewhere, I'm sure!
> Sarah, an avid gamer, seemed to recognize this strategy immediately.
As soon as I read this, I immediately thought of the puzzle in the last gym of the Gen 3 Pokémon games (Ruby/Sapphire/Emerald). It was a big ice platform that cracked after stepping on it once, and backtracking meant falling through the ice and having to start all over again.
It made me think of the “step on all the colored tiles without crossing your path” puzzles in Links Awakening!
I wonder how the optimal path changes if turns are penalized as well as steps - in real life turning a lawnmower takes a lot more effort than moving the same spot twice in a straight line.
A lawn mower cuts grass whether you push it forward or pull it backwards: my trick when I was a kid was just to push and pull it back and forth across the lawn without changing the direction it faced at all.
Seems like the most optimal path would also account for returning the mower to its storage space...
This is a pretty big deal once you scale it up.
https://link.springer.com/article/10.1007/s11119-023-10033-9
Hear me out: It's a low flying drone with a propeller on the undercarriage as well which is used to cut the grass. Voilá, no wheel lines.
You mean a flymo. It's more hovercraft than drone, and much older technology. Works reasonably well on non-lumpy lawns - my neighbor had one in the 1980s.
www.flymo.com
Fun article, but this didn't actually answer the "why" question for me...
The author's objective function is incorrect, negating the core claim.
That reminded me a lot more of Chip's Challenge than of mowing a lawn. Just missing some blocks to push!
have to say i really enjoyed how this was presented.
no comment on the results or the premise of are people good at mowing or just good at playing "snake"
i would be very curious to see if this same result held up when you also factor things like turn effort or "prioritize the longest straight lines bc they look cool"
I'm sure I'm the odd ball, but ever since I was a kid, I liked the diagonal patterns much more than the square lines.. and I'm not just talking about lawns. I would vacuum my room on diagonals.. lol
When I mow lawns I teleport forward 6ft at a time and then a 6x6 square of grass vanishes beneath me.
In the nude.
Better is subjective here i think, what about laying some nice stripes or diamonds? How do you work in some aesthetics and still stay efficient?
it's Q*bert!
Mowing the lawn is never about mowing but people wanted to be busy, feel productive for a bit, some exercise, and helping to clear thoughts. For others it might be driving long roads, or doing chores inside the house, cycling, and such. Otherwise, it’s inefficient and useless, the lawn can be replaced with better flowers or even small garden, or simply pave it and save the time and efforts and electricity/water doing a repetitive thing like it, and also spare your neighbors the noise 7am Saturday morning, and if you look at the bigger picture, the process is a total waste, you waste water to grow crops only to waste electricity to mow it later, it provides no useful product or outcome either, no flowers, no food, not even cosmetics like fancy looking garden, plus the maintenance and/or initial installations.
It makes like 30 turns as the "best", but if the program ever used a mower, it would want to minimize turns.
Probably the same reason some food delivery people place my bag next to the door instead of in front of the outward opening door where it will tip over, unsure what that reason is but I’d like to know.
The optimum path does not take into account row striping :)
100 percent first try. Too easy.
I hate this game.
Mowing a lawn is about creating straight even lines not optimal routes.
Dig that lawn out and start growing food, ffs.
Why? My dogs and I get much more enjoyment out of a lawn than we would a garden. I can buy food, but space to play fetch is a premium.
I only have a front lawn (townhouse) and I'm not allowed to plant whatever I want in it.
If somebody reports it, the city can and will come tear it out and charge me for the work.
"Ownership" is a funny thing.
I can only fell 10% of my trees per year even though I have a felling licence...and why do I need a licence?!
Grass is food. If you don't have animals out there grazing it, that's where you've gone wrong.
Or just eat the grass?
You first.
I have 45 acres of grass and woodland (some is growing food), not a lawn in sight lol.
the article kinda ignores that the direction of mowing matters, optimal path would look shit
> You’ve probably done something like mow a lawn or vacuum a rug hundreds of times without thinking much about it. Some part of your brain works out a route that’s usually good enough to get the job done. People are actually pretty good at this, and better than you’d guess for a problem that is hard for computers. We wanted to see it for ourselves, so we built a lawn and asked people to mow it.
Is there supposed to be more content than this?
... If JavaScript is required to view more content than this, why?
> ... If JavaScript is required to view more content than this, why?
By "asked people to mow it" they're talking about the visitors to the site. It's interactive.
It's somewhat a mistake not to have "skip to results" in the actual html. On the other hand, it makes sense they want to force people to see the landing page first. And unless they want to add a whole rendered fallback the results page isn't going to be super functional without javascript.
Without JavaScript, there is literally no page text beyond that point until the footer. Thee is no indication that there would be any interactive content. There is also no discussion of what the interactive content demonstrates after it. There are no "results" to which a "skip to results" could skip.
Yeah I know. But "literally no page text" is only missing a couple words. The correct version is super short.
The noscript version needs a button and a big box with an X. And even then the button would only be half-functional so I understand skipping it. It mostly just needs a "missing content" warning.
We get it, you're really proud that you choose to disable JavaScript, and need to let everyone know about it