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
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.
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.
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.
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.
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.)
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?
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.
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).
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.
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 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. So yes, after decades of indifferent mowing, one can detect where wheels have been moving topsoil.
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.
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.
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.
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?
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.
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.
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!
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
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"
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
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.
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'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.
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.
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.
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.)
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.
Why not mow in a spiral then hit the 4 corners last?
The lines change mowing so much!
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?
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.
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).
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.
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.
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. So yes, after decades of indifferent mowing, one can detect where wheels have been moving topsoil.
"Worked as a landscaper during my youth"
> 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.
>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.
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.
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
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.
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.
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.
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.
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!
That reminded me a lot more of Chip's Challenge than of mowing a lawn. Just missing some blocks to push!
This is a pretty big deal once you scale it up.
https://link.springer.com/article/10.1007/s11119-023-10033-9
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
The author's objective function is incorrect, negating the core claim.
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"
When I mow lawns I teleport forward 6ft at a time and then a 6x6 square of grass vanishes beneath me.
In the nude.
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 :)
I hate this game.
Mowing a lawn is about creating straight even lines not optimal routes.
100 percent first try. Too easy.
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.
Or just eat the grass?
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.