The masterstroke of this design is that it creates a time vortex when the windspeed reaches 88.5 miles per hour, sending it back to 1997 to crush Bill Clinton to death so that his successor ratifies the Ottawa Treaty banning landmine sales from the United States of America.
Thousands of lives saved by such simple engineering!
This isn't to be used alone for clearing minefields - as it's not going to get 100% coverage.
It's still a useful tool for surveying minefields or perhaps doing a cheap first-pass before sending in pricey equipment
This is a great solution, but it seems like it is designed to follow a random path (that is, where the wind blows). How do you prevent this from creating a false sense of security, when in fact there are areas of the desert which have not been tested?
I remember when I was little I built a small platform high up in a tree. My father refused to let me put a rail around it, because the sort of rail I could build would not be enough to stop a fall and would only encourage a sense of complacency.
Of course, I do not mean this as a criticism of the design, but merely as a path for future exploration.
A truly random path would be pretty good, since you approach perfect coverage with repeated runs. Alas, the wind direction is not random, particularly in this kind of environment.
An interesting demining approach is to buy the local villagers lots of sheep and pen them in the areas of concern. Once in a while a sheep goes boom. Meanwhile: wool and baby sheep as a side effect.
Truly random paths would give you the 97.8% probability that there are no mines left, over a course of N field traversals. Obviously, that ain't good enough.
Also, it suffers from the same problem as the Mine Kafon, in that it's a more-or-less random walk and you won't know whether the whole area is cleared.
Just to be clear, are you saying that it is only very slightly better for a sheep, rather than a human child, to be maimed by a land mine? That seems to be what you wrote, but I'm having a very difficult time reconciling your world view with my own, so I'm hoping that I am somehow misunderstanding.
You're intending to be counter-intuitively insightful, but there are few scenarios in which a sheep is more valuable than a child. You could sell a child for more than you could sell a sheep. The children who are in the most danger from mines are past infancy, and that is when children begin to be worth more than they cost to maintain if you aren't sending them to school.
Wilfully inflicting a land mine injury on another creature? Sheep might not be as smart as people, but I bet they feel pain every bit as much as you do or a child does.
And given that, what does it say about you as a person that you'd think that it was a good idea?
Both involve sentient beings. This is not about valuing children less. This is about valuing sheep more.
This might be an uncomfortable world view, taking how much we depend on non-human abuse in our civilization, but it's not the only case where we built it on abuse of others, human or not.
You do realize these sheep are not pets and are not treated as such in Afghanistan, right? They will be used for milk, wool, and some will be slaughtered for meat.
This makes sense for the people who live there- the dessert is not an easy place to grow crops and a multi-purpose animal is very valuable for survival purposes. Animal rights is a tough sell in a place like this.
It seems like an ineffective non-systematic way to de-mine that damages animals that the Afghans value though.
Of course — I said dependance. Just as we depend on abuse of countless people, and we can't stop either overnight. It doesn't change the fact that it's horrible one bit.
If it was neccessary to save my child, I'd eagerly murder every sheep in the world in the most painful way possible with my hands.
Animal lives and welfare do matter, but when it comes to weighing human lives against animal lives, only a ratio of something like a million-to-one may be ethically justified. Definitely not one-to-one sheep to a kid. Definitely not even hundred-to-one. It could be reasonable to sacrifice a few humans to spare a species from extinction, but not to save some individual animals, however sentient they may be.
My fellow humans are important. Leaving a kid to die to spare a hundred sheep is disgraceful.
Wow, it must be really scary to be a kid of someone with that kind of violent fantasies.
Now, as a mental exercise, I suggest you replace "animal" with a racial slur of choice. Because, seriously, 100 years ago people said that about all kinds of other people. They actually still do. Having a slightly higher amount of common DNA is not a justification of whatever-the-hell.
And at the point where you say "however sentient they may be", you just go into downright I-don't-want-to-meet-that-person-in-a-dark-alley territory.
The first sentence was an overexagerration, but I have had near-death experience for my kid, and that makes you reconsider various things in life seriously. I am certainly not a violent person whatsoever, but from that incident I understood that I would disregard almost everything if stakes were that high once more.
But, coming back to the original issue - it all comes down to two different theories of ethics, opinions about what is good and just: in essence, if some choice equals to murdering one to save ten others [a simplification for the sake of argument, assume that no alternatives exist] , would doing so be good or evil?
There are valid arguments to say that doing so would be evil by the nature of the act.
There are valid arguments to say that not doing so would be evil, by the result of the act (10 others dying instead of 1).
And I do value a human life of "a racial slur of choice" more than a hundred sheep - archaic laws did value human lives in a couple dozen of domestic animals, and that whas when a peasant's life was considered cheap; we can and should be better than that.
And I do value my kids life more than your life - that is also quite natural, I doubt that you'd find any parents who think otherwise.
It all adds up. I mean, if you start discussing life and death, then most solutions aren't pretty; and discussing only 'pure' solutions means not saving (= murdering) lives - such as the lives of people that could have been saved by some sheep.
How about we don't murder any sheep or children, and instead spend the money that we would've spent on sheep torture on better fences, signs, education and mine clearance? While you're at it, you can lobby your elected representative to spend more than the currently pitiful amount that they do on mine clearance.
ps. A million-to-one? You have a very interesting idea of "ethics". Also of "definitely".
How are people supposed to shear those sheep, or take care of them during lambing/when they're sick, or to feed or water them, if they can't actually enter the pen? It seems to me that sheep penned in a land mine field would either be useless, because you couldn't go near them, or worse than useless, because they'd be a temptation to enter the dangerous area.
No, you're right, I never worked on a farm; I simply grew up in a farm.
Sheep are scared stomachs on legs. When they see their daily source of food and nothing suspicious (such as a stranger), they reach. My grandfather could also call them and they would reach even with a stranger present.
The point is that these people have sheep and goats already for basic subsistence. They already use these dangerous fields. By buying them more animals instead of (or as well as) spending money on expensive technical demining kit, you increase the probability of mine clearance a bit, while making the person wealthier. Sheep reproduce and excess lambs can be sold, making the process sustainable.
Similarly, training locals in demining with donated basic equipment is also good, in the sense that these new experts can sell their services locally. The down side is that these brave people take extra risk.
Both the livestock and local training methods were used in Bosnia, which had a serious landmine problem recently.
Sheep like to eat stuff like grass, which does not appear to be plentiful in the areas in question. You could feed hay, but then the sheep will congregate around the area where you dump the hay instead of wandering around getting blown up. Not to mention that hay (as well as fencing, and possibly water in some of those areas) would be expensive.
Multiple types of animals have been trained (EG http://www.cnn.com/2010/WORLD/africa/09/07/herorats.detect.l...) to detect land mines. Most animals have much more highly developed senses than ours. It is possible that the flocks of sheep would learn over time to avoid landmines. It's possibly they already do this. It would be better to teach the local people how to train animals than to damage their valuable livestock.
One alternative to this is to just create lanes with rope.
i.e. you drive a stake in the ground on either side of the field (if that's even possible) and just create a lane where each of these rollers roll through.
It won't work in every single landmine, but it can provide a 'grid' within which you know what has been cleared and what has not.
You don't have to run them through the minefield really. You could potentially start from the side and drag two ends of the lines from the side.
May be harder with heavy / long lines / wind, but come on - we can build hanging bridges over very high canyons without anyone levitating to the other side first ;)
These things look rad. If there was some way to keep track of the landmines detonated per Mine Kafo I could totally see making a game out of this. Donate $10 and get your name on a rolling mine clearer. Live scoreboard keeps track of mines cleared!
Mine rollers and mine flails like this have been tried and tested since WWI but none have proven completely effective in finding and clearing mines. This is partially because they only work well on totally flat terrain and rapidly lose effectiveness in rougher terrain where a large number of mines are typically buried.
In practice, solutions like this achieve only 50-60% effectiveness at clearing minefields, which makes them useless for civilian demining which demands a 99% clearance rate.
For that reason there's been a ton of research in this area which has resulted in better demining vehicles and interesting new techniques such as using sniffer dogs or rats to detect the explosives inside landmines. This is especially useful for the numerous plastic-shelled landmines that resist standard detection methods.
Using a layered approach with these techniques, civilian demining organisations like Mechem[0] (which pioneered the use of sniffer dogs) are now able to achieve a high enough clearance rate to make areas safe, though the work is expensive and time-consuming. If you support this sort of work, donating money to demining NGOs would be better than funding yet another ineffective mine roller.
The real question is it any cheaper to de-mine 100m^2 with 50 land mines in it, than it is to de-mine the same area with 100. My expectation would say no.
This method won't make the land usable safely, so the reality would see a bunch of cheap partial clearances with no real benefit - the areas will still be no go zones.
This is a topic I know next to nothing about, but I have two responses for your question:
1. Is it really useful at all to only clear half the landmines? I would assume the area is still too dangerous to be useful even with only 40% of the landmines remaining.
2. Given #1, it would follow that additional (expensive) land mine clearing technologies would always be needed in addition. I don't know how these work or are billed out, but if the cost is per square mile or some similar area-based unit (instead of say, per mine), then clearing out 60% of the landmines first would not affect the supplemental cost. You've still got to cover the same area to catch the other 40%.
Anyway, you've had my disclaimer at the top, and I could be totally wrong about this :) Just some thoughts.
I'd expect the there would be a search cost, a removal cost and making safe cost. That the search cost would be lower than the removal cost (less skilled, less dangerous).
I'd be extremely surprised to find that a commercial clearing of mines for an area with 40 mines was the same cost as clearing that same area with 400 mines. Equally that clearing a double-sized area with 40 mines was the same cost.
Either way you'd reduce costs - reducing search area or reducing removal+disposals needed. [fwiw _djo_ intimated that the increased cost to the process of debris removed any utility such devices might have but also suggested that similar devices are in current commercial use].
With a sandy area I'm wondering how well a dragged scarifier like implement would fair.
That's not a bad question. Thing is the organisations that perform thorough civilian demining already start off with cheaper methods such as mine rollers to get some of the mines, and the ones they use are arguably cheaper to use because they're strong enough to require fewer replacement parts. For instance Mechem has a CASSPIR armoured vehicle with steel wheels that'll happily trundle over anti-personnel mines all day without needing repairs (though anti-tank mines necessitate a wheel change).
And as another poster (codebeaker) pointed out, it's also far worse for demining teams to have random shrapnel spread over a minefield, as it makes the job of identifying mines much more difficult. That's one of the reasons why demining teams use systems that are less likely to break and leave pieces on the minefield.
So it's unlikely that this product will save any money.
The difference is this costs 40 € and can clear a couple of mines for that money. Afghanistan is not exactly rolling in cash, so any cheap improvement is a valuable addition to the toolbox - it provides a way for poor communities to do something while they're waiting for America to donate enough money to clean up their mess.
My guess is that demining NGOs will be using something like this soon enough if it proves effective, though, so any donation dollar really is best sent there for maximum effectiveness.
Demining organisations already use mine rollers in situations where they make sense. As I said, they use a layered approach, with mine-rollers, vehicles, sniffer dogs, etc used as the circumstances require it.
The problem with communities doing it themselves is it'll give them a false sense of security about the safety of known minefields. This could potentially increase the number of mine-related injuries because far more people will attempt to live in or traverse known minefields which they wrongly believe to have been cleared yet which still contain roughly half of the original mines.
I think you went hyperbolic in your GP post. You said that mine-rollers were "useless" - which they clearly are not if they can clear mines as a first pass (yes given the right terrain).
I disagree. I said 'solutions like this', i.e. mine rollers as a single technique, were useless. I stand by that statement because the stats are quite clear on this as even in ideal conditions (which are seldom present) a roller can only be expected to detonate half the mines in an area. In fact in many areas rollers can't be used at all because of the terrain and vegetation, so demining teams use a combination of other techniques.
This concept is actually worse than useless because of the debris it scatters across a minefield and the false confidence it provides. While I applaud the designer's attempt to create something useful, his lack of domain knowledge has resulted in a design that doesn't solve the problem and may make things worse.
If I come across as bit hyperbolic, it's because I strongly believe that this is a horrible and underreported problem which demands both carefully-considered and effective solutions and acknowledgement of the people whose hard work and sacrifices have made so many places safe again.
It seems silly to laud an ineffective design while there are people out there every day clearing landmines with solutions that work.
You're saying something akin to: 'tyres are useless because cars also need engines; people are already effectively using tyres; often tyres don't work'.
>It seems silly to laud an ineffective design while there are people out there every day clearing landmines with solutions that work. //
Solutions that according to you - strongly implied eve within the parent comment - include rollers.
So
"Mine rollers are useless [for that purpose] and so should never be used to clear mines under any circumstances."
I find this debate a bit perplexing: You're asking me to answer a straw man argument that nobody here has made, so I'm not entirely sure what your overall point is.
Mine rollers can be effective for civilian demining, but only in the right type of terrain with the right type of vegetation and when used in concert with other demining techniques. I didn't just imply that it my posts, I stated it outright when asked when and where mine rollers could be effective.
But when used independently, as the Mine Kafon's video shows, mine rollers are useless for civilian demining and, because of the false sense of security they provide, could even be harmful. A minefield that people believe to be cleared, but isn't, is far more dangerous than a minefield that's demarcated and which most people know to avoid.
So I'd amend your provided phrase and say:
"Mine rollers used independently are useless for civilian demining and should not be seen as civilian mine clearance solutions."
You've really squirmed and squeezed to ensure you can still assert the uselessness of the technique whilst still maintaining the counterpoint of it being used and effective; well played.
"Mine rollers can be effective for civilian demining".
So, hang on maybe they're not useless as _djo_ was contending. But wait, why is it they can be effective we ask, oh right it's because they
The context of those phrases, which you removed, made it clear enough that the first phrase referred to mine rolling as part of a system of complementary techniques while the second referred to it as a technique used on its own.
By intentionally removing that context so as to attempt to prove the phrases contradictory, you've proven only that you're more interesting in trolling than in debating this seriously. So I see no point in continuing this conversation.
>you've proven only that you're more interesting in trolling than in debating this seriously //
The point was that you started by saying the thing was useless and persisted to try and defend that point despite acceding that similar devices are used and useful.
There was nothing to debate, I simply called you on over-exaggeration and you refuse to accept it is. Fair enough you clearly know more about the domain in question. But the information you've given is the counterpoint to your initial claim.
The reason I quoted without the full context was because that indicated your initial argument. The extended quote was how you adapted your argument against the claim that you were wrongly portraying the true position.
Afghanistan is not exactly flat land and mines are not exactly placed in the open field and this tool is not exactly as heuristic as my robot vacuum cleaner and this is not exactly anywhere a valuable addition because even after 100 years you cannot be sure they cleaned an area 100 percent.
I was an SFOR-soldier in Yugoslavia and I saw several crippled kids sitting next to a football field watching other children play in tears while others were reading superman comics which informed them about not touching mines and how they look like. They deserve hope and not a joke ..
The mine roller looked to be made of three parts: the black core (expensive); the plastic plates (very cheap); and the bamboo poles (cheap as shit).
Provided the weight came mostly from the core and the core never itself got blown up, I'd saw it can clear more than a couple of land mines for the money. Perhaps, a land mine per plastic plate, and if an NGO is paying for Afghani wages, you can roll that baby all day.
Bonus: it creates a relatively "safe" path in its footstep.
Regrettably the only "safe" place is the location where any mine actually exploded, couple of feet away, not so much.
The haphazard nature of the exploration is the big issue and using this would most likely give people a false sense of security. This is an interesting piece of lo-tech design created with the very best of intentions, but it is essentially useless and may even be harmful.
The point is that clearing a couple of mines is not useful and is even harmful.
A patch of land is unusable while mined; useful when cleared, but if it's 50% cleared (or 80% cleared) then it is murderous, because then it gets used, resulting in dead and crippled locals.
>The point is that clearing a couple of mines is not useful and is even harmful. //
A field that you must use to avoid starving has 100 mines in it, you can remove [on average] 55 of those mines using a locally made and affordable device. The field will be used anyway. Would you deploy the device?
You're leading a demining team. A village have asked you to clear a field that they rely on for crops (other areas are too rocky or haven't enough good soil). You can send out some locally produced cheap demining devices first that will remove about 50% of the mines reducing the risk to your team. Would you deploy the device?
"is not useful" you say?
Your position appears to be a counterpoint to a statement that these devices will clear all mines effectively from all areas in all conditions. I've not seen anyone make such a claim. Your position is only tenable when a mined area is not needed by the locals.
The devices, and the beautifully made film, have another purpose which is far from useless too - attracting attention and thence funding for demining projects.
Your argument hinges on the statement "The field will be used anyway." As people are not binary in their decision making. The arguments of others that partial demining will increase the likely hood of use and thus increase the risk of harm is completely valid.
I'm responding to a universal claim by indicating a single instance in which the claim would be wrong, falsification basically. My argument is that those saying this is useless appear to be going against evidence that it has - whilst limited - genuine utility.
The argument for partial demining leading to increased harm requires that land that is mined remain completely unused otherwise or be completely demined. As soon as there is any use then reducing the number of mines statistically [and yes maybe simplistically] will be better - which would you rather, walk across a field of 200 mines or 400 mines?
Professionally demined areas are only cleared to a level of about 99.6% according to the thread so the argument for increased risk of harm still works in that case - more risk from mines when walking in a 'completely' demined area than in an area in which mines have never been placed.
>I'm responding to a universal claim by indicating a single instance in which the claim would be wrong, falsification basically. My argument is that those saying this is useless appear to be going against evidence that it has - whilst limited - genuine utility.
Congratulations, you have imagined an instance where the this device can be used. The constraints you have imposed are: that the land comprising the minefield must be used and will be used regardless of whether the mines are cleared; that there is no predetermined upper bound of what constitutes acceptable civilian casualties; that the cost of subsequent de-mining is immaterial (e.g., when the local population decides that maybe 50% wasn't good enough after all); that the terrain is suitably flat for the Kafon; and that the extant vegetation will not interfere with it.
Now, does the scenario you imagined actually exist in the real world? Or is this purely an exercise in mental masturbation? I would suggest that the scenario you imagine is not one that is common in the real world. Afghanistan is not the first place where mines have been used, and yet there don't seem to be many farmers in Vietnam or Cambodia re-purposing known, uncleared minefields. Almost all civilian mine casualties seem to be the result of people using areas where they thought there were no mines. My take-away from _djo_'s comments is that the Mine Kafon primary effect will be to make it very cheap and very easy to create large areas where people believe there are "no" mines, and that is a problem.
In engineering it is often true that a general solution is preferred to one which only works in special cases. The Mine Kafon seems to be, when viewed in the most charitable light, a very special-purpose solution which adds no value above the general solution. Viewed in a more balanced way, the Mine Kafon is a solution in search of a problem[1]. User _djo_ seems to be saying that, there are already tools which do what the Mine Kafon does, and they minimize not just the economic cost to the de-mining organization but also the human cost to the local population.
[1] Do you recall the article several days ago by Paul Graham about how to find good start-up ideas? One thing I took away from that discussion is that you should try to find solutions to existing problems instead of looking for problems that fit your existing solution.
Turns out that the Dutch armed forces agree[1] that it's not so easy either:
"The Mine Kafon has just been tested by the Dutch Explosive Ordnance Disposal Unit, which has concluded that it is not suitable for mine clearance (which requires a more systematic approach), but at $50 each could be used as a cheap and safe way to identify dangerous areas that need demining."
Not suitable - for military purposes. DEODU is a military organization, and any conclusions it makes must be viewed in that light. Mine-clearing is a war tool as much as it is a peace tool.
You're getting downvoted because civilian requirements are higher than military.
If you manage to clear a field such that there is a single mine in it, it will likely become safer to march through than avoid - there are plenty of other things that can kill you, and they must be triaged in war time.
However, if you take that same field after the war is over and put a farmer on it, it's basically a matter of time until he steps on it. His only chance is that an animal blows it up, or the mine is defective.
Therefore, if it's not suitable for military clearing, it is even less suitable for civilian purposes (except for figuring out what needs to be cleared, as they say).
I upvoted you and your parent. His assertion triggered your informative response, which I'd have had to do without otherwise. He wasn't annoying, just wrong, which is entirely acceptable as long as we are civil about it.
While you are probably right about this mine clearing tool, there have been previous examples that show the military’s requirements for particular items are much higher than civilian requirements in times of relative peace. Obviously when there is a world war, requirements go down dramatically to get things done, but when making purchase decisions in time of peace it seems that requirements are very high. Items like devices/vehicles etc. get very expensive and complex.
[Added] I’m talking from a US perspective, but it seems plausible that most developed country militaries spend on technology when in time of peace.
If you watch the video, that's all that this is claimed to do. The headline it was posted under on is.gd and here on HN claim it's for clearing mines; but at 2:50 he says "My ideal scenario would be to provide it for local people as a tool to cheaply survey this danger in their own environment."
All this discussion is because of an over-enthusiastic headline, not what the tool is actually intended for.
I can't verify since I'm at work, but I'm pretty sure he talks about how a single roller can clear up to four mines. To me that implies this is meant as a tool for clearing the mines.
Surprise, surprise .. 'authorities' are 'doing it better'.
Well, this tech is not for the authorities. Its for children to use to help clear the danger away from their villages. Its a tech that anyone can apply, not just those with access industrialized manufacturing facilities and the 'mine-clearing economy'.
99% clearance rate? I think most villages would be happy with being able to clear a safe path for their kids to walk on, at any rate. This cheap, useful, easy to apply technology allows that to occur. Make 100 of them, let them loose for a few weeks, and suddenly 99% seems a lot more doable with this method than any other - especially for the resource-lacking Afghani villages who are left with the war liability that the imperial, authoritarian states, have given them ..
What you don't understand is that mined areas are typically marked off, which is why it's children who are most often affected because they're more likely to ignore warning signs. Nonetheless because most people stay away from those areas the risk is lower.
When you demine an area, you signal that it's safe for the population to move in, which means that if you haven't actually cleared 99% of the mines you're increasing the possibility of a person stepping on a mine. That's the reason the UN demands such a high clearance rate before an area can be declared safe for humans to live in.
A safe path is 99+% clear, this is not a 99% solution. And please keep in mind human nature, a false sense of security will kill you. If it is a flat out no go zone people will avoid it, if they think it is safe they will find mines on there own. Fear can keep you alive and anything that lessens the fear of the minefield without rendering it safe is a net loss of human life in my opinion.
I agree. Demining for human/civilian is harder than just to clear a way for a military purpose. The thing with projects like this roller is that it actually takes money and time from the things that really work, like sniffer dogs.
My favourite solution so far was to use plants that turn red near mines, thus pretty much creating red circles around their location. I read about it few years ago in a science magazine, but somehow the research seems to have died.
Unfortunately Aresa Biodetection, the company behind that, stopped all research into their strain of the thale cress plant for landmine detection. I would guess that they hit a dead-end in their research, perhaps they could not make it reliable enough.
And how do you irrigate and fertilize mined territory in the Afghan desert in a way that's cheaper than actual demining? Walking around with a hose or setting up an irrigation system seems too dangerous (or require partial demining to create footpaths anyway), and using any type of aircraft is expensive.
There are already large numbers of aircraft flying over Afghanistan looking for the Taliban, aren't there? That's obviously not too expensive, it's just a question of political will.
> civilian demining which demands a 99% clearance rate.
Is that an actual official classification? It seems a bit low. "There were 1000 mines here, and we found 990 of them. Good enough, I say. Move in, and bring your kids!"
The actual figure is 99.6%.[0] That's the statistical confidence level for clearance required of contractors and to declare an area safe, because it's impossible to prove beyond any doubt that there is no longer a single mine left in an area. Demining teams still aim for 100% though.
For instance, Mechem's sniffer dogs have displayed a 90% success rate on target filters, so statistically having three dogs per team perform cross-checking paths on the field should produce the 99.6% confidence level for the type of mines the dogs can detect.
>For instance, Mechem's sniffer dogs have displayed a 90% success rate on target filters, so statistically having three dogs per team perform cross-checking paths on the field should produce the 99.6% confidence level for the type of mines the dogs can detect.
I know nothing at all about this area, but presumably this only works if the 10% failure rate is random rather than based on some quality of the undetected mines?
More generally, that the chance of each dog X failing to detect each mine Y is independent. If it is independent, then you get the (1-P)^3 chance of failure, meaning 0.1% for all three failures if each dog has a 90% chance of succeeding.
But of course that isn't true in reality. There will be properties of the mines (burial depth, ground cover, degree of leakage of sniffable explosive material) that affect its chance of being found by all dogs searching for a mine.
If you wanted the parent poster's statistical chance, you'd want three 90% ways of detecting mines that are truly independent. Perhaps a dog, and infrared imaging sensitive enough to detect the explosive material (I have no idea if that's possible), and this article's mechanism.
Correct, that's why in real world scenarios companies like Mechem will use a number of techniques. The link I posted earlier makes it clear that the UN considers a combination of techniques necessary to meet the required standard.
_djo_ I like a lot of what you have to say. I'd also like to know more details around why rougher terrains are a problem? As an idea .. say you had a number of these things, small and heavy with a cheap protected motor and gps guided ... would that have a chance of better clearance? more in the 99% range?
I think the solution mentioned in the clip might indicate a starting point. no?
A huge problem is that land mines get unreliable over time and thus a mine flail might hit it but it does not detonate. However it could still go off when handled later.
That's an anti-tank mine. All anti-tank mines are designed to not blow up when people walk on them, only when heavy vehicles drive over them. We're talking about anti-personel mines here.
This story is only about the short film made about the Mine Kafon. If you want more info on the deminer itself you should check out the designer's blog. His concept also include GPS tracking and an online platform, as suggested here on HN.
How about a bio-solution? A fast-growing 1-year non-reproductive plant or weed that's engineered to grows best (or not grow) around old explosive chemicals? I imagine dropping seeds from a plane around a (old) minefield. After a few months you'll see plants popping up and you'll know that's where the old mines are.
Yes! "Plants... are said to turn from green to red when they come in contact with explosives in the soil. " Read it some time ago and thought it was brilliant.
This creates an incentive to clear only first few mines, because they will become progressively harder to find and an hourly reward rate would quickly fall to near zero.
In an area where £10 might be enough money to pay for a doctor's visit for a deathly ill child, or feed a family for a month, you'd probably get a lot of people risking their life and getting killed to get their hands on just one landmine, which strikes me as a singularly inefficient and cruel way to try and clear those fields.
Sorry for going completely off-topic, but I'm slightly annoyed at the overuse of adverbs like "hauntingly" and "wildly". "Hauntingly" might be tolerated here even if it is very hyperbolic, but "wildly low-cost"? What does that even mean?! Why can't the headline just be "Beautiful, Low-cost Solution to Clearing Afghan Landmines"?
Oh, and another one that has been popping up a lot around here lately is "vanishingly" (I've noticed Patio11 likes it a lot!).
This title is feels like a product of the previous article "how to pick your titles using AdWords" it is just missing "1 cheap trick that an Afghan child came up with that makes demining companies furious"
I believe that studies (which I cannot find citations for now) determined that this device makes the sutation worse, as it scatters metalic debris over a wider area, and that human mine clearance teams have to work an order of magnitude harder, cleaning up tiny bits of shrapnel. Before an area can be marked clear, it has to be devoid of all unidentified metalic shards, and of course, all mines of other materials. As a result, it was the determination of more than one agency that this device makes things worse.
With that said, props to the designer for working on a solution to a difficult problem, but perhaps more domain knowledge would have lead him to a more suitable solution
Some of my friends served in the IDF engineering corps and cleared mines in the Golan as part of their service, I assure you they're not goats :)
(Many mines are also designed to be triggered under a certain weight, so goats can walk freely on them and not get hurt)
One interesting method I read about is seeding the area (from the air) with a certain flower. When it blooms, the contents of the ground affect the color, so on an aerial photo you could locate the mined areas.
That looks pretty cool. I don't think it's correct to advertise it as "clearing" minefields though. It looks like a useful tool for identifying them, and making them a bit less miney.
My understanding is that manual, low-tech demining is not actually that dangerous. It requires some training and caution, but generally speaking anti-personnel mines aren't likely to do you much damage if approached carefully from the right angle.
Whereas the approach of getting machines to jump up and down randomly is expensive, likely to lead to a false sense of security and may even cause mines to get pushed into positions that make them more dangerous.
You could get a lot of Afghan workers enthusiastically doing low-tech demining for not that much money. Every now and again one of them would lose a finger or two but maybe that's not such a bad outcome considering the alternatives. People tend to get much more excited about expensive, higher tech solutions because they're sexier and the people who develop them (often the same people that made the mines in the first place) are better at selling them.
In this flat terrain it will be much more simple and efficient just to drive tanks in strict patterns and blow the stuff below them.
Or you could just create a fleet of cheap ground stomping robots with armored feet designed to deflect the blast wave that are powered by cable and remotely controlled.
One of the things which terrified me most about landmines was seeing signs all over pointing out that it was a violation of UCMJ and huge fine/jail/etc. for stealing land mine marker signs. I always assumed minefields were all either professionally laid (and marked) or had been marked by subsequent forces (and thus could be avoided), but I hadn't realized anyone would be stupid enough to steal the markers.
In practice the best way to clear most mines (to a military standard, which is far lower than civilian) seems to be explosively generated overpressure -- they launch a bunch of explosives, usually a linear charge, in the direction of travel, and it drops and blows up, clearing a path.
Using trained rats to clear mines (cheaper than dogs) seems to be the best upcoming way to meet the civilian standard. It's labor intensive, too, but most countries with landmines have relatively cheap labor, and the training to be a rat operator isn't as difficult as to be a full EOD tech. It's definitely one of the charities worth supporting.
Is there a cheap way to add a chip to help map the precise route taken? It would increase the cost but I'd think it would help to know that an area has been covered several times.
great case study of why product design is very hard and requires deep understanding of the factors involved.
as others have pointed out:
- your goal is to clear an area, hence you need a systematic approach to be able to deem a strip of land safe. random paths do not help
- this identifies and detonates in one go. but if the mine does not get triggered, the identifier is lost. these are two separate tasks. merging them would only work with 100% detonation rate.
- this does not work against other types of explosives. anti-vehicle mines, unexploded ordnance, cluster ammo,...
the upside is definitely the low cost. the low effectiveness rate and unreliable pathing makes it lossy though. which with mines is not acceptable.
When you put effort into promotional video, how hard is it to Google the actual Russian text for "danger minefield" rather than put up a shield with random Cyrillic alphabet soup?
"Hauntingly Beautiful" ... It's interesting to me how the feet on each leg is designed to catch the wind. You might not get the torque needed to generate electricity, but the gentle continuous "nudge" needed to roll across the desert can come from any direction.
180 comments
[ 6.0 ms ] story [ 220 ms ] threadThousands of lives saved by such simple engineering!
I wonder if there is a way to make a more cylindrical version of this, which might follow a more predictable and controlled path through a minefield.
This isn't to be used alone for clearing minefields - as it's not going to get 100% coverage. It's still a useful tool for surveying minefields or perhaps doing a cheap first-pass before sending in pricey equipment
I remember when I was little I built a small platform high up in a tree. My father refused to let me put a rail around it, because the sort of rail I could build would not be enough to stop a fall and would only encourage a sense of complacency.
Of course, I do not mean this as a criticism of the design, but merely as a path for future exploration.
An interesting demining approach is to buy the local villagers lots of sheep and pen them in the areas of concern. Once in a while a sheep goes boom. Meanwhile: wool and baby sheep as a side effect.
That might kill sheep outright, but I doubt it.
Also, it suffers from the same problem as the Mine Kafon, in that it's a more-or-less random walk and you won't know whether the whole area is cleared.
Just to be clear, are you saying that it is only very slightly better for a sheep, rather than a human child, to be maimed by a land mine? That seems to be what you wrote, but I'm having a very difficult time reconciling your world view with my own, so I'm hoping that I am somehow misunderstanding.
And given that, what does it say about you as a person that you'd think that it was a good idea?
This might be an uncomfortable world view, taking how much we depend on non-human abuse in our civilization, but it's not the only case where we built it on abuse of others, human or not.
This makes sense for the people who live there- the dessert is not an easy place to grow crops and a multi-purpose animal is very valuable for survival purposes. Animal rights is a tough sell in a place like this.
It seems like an ineffective non-systematic way to de-mine that damages animals that the Afghans value though.
Animal lives and welfare do matter, but when it comes to weighing human lives against animal lives, only a ratio of something like a million-to-one may be ethically justified. Definitely not one-to-one sheep to a kid. Definitely not even hundred-to-one. It could be reasonable to sacrifice a few humans to spare a species from extinction, but not to save some individual animals, however sentient they may be.
My fellow humans are important. Leaving a kid to die to spare a hundred sheep is disgraceful.
Now, as a mental exercise, I suggest you replace "animal" with a racial slur of choice. Because, seriously, 100 years ago people said that about all kinds of other people. They actually still do. Having a slightly higher amount of common DNA is not a justification of whatever-the-hell.
And at the point where you say "however sentient they may be", you just go into downright I-don't-want-to-meet-that-person-in-a-dark-alley territory.
But, coming back to the original issue - it all comes down to two different theories of ethics, opinions about what is good and just: in essence, if some choice equals to murdering one to save ten others [a simplification for the sake of argument, assume that no alternatives exist] , would doing so be good or evil?
There are valid arguments to say that doing so would be evil by the nature of the act.
There are valid arguments to say that not doing so would be evil, by the result of the act (10 others dying instead of 1).
And I do value a human life of "a racial slur of choice" more than a hundred sheep - archaic laws did value human lives in a couple dozen of domestic animals, and that whas when a peasant's life was considered cheap; we can and should be better than that.
And I do value my kids life more than your life - that is also quite natural, I doubt that you'd find any parents who think otherwise.
It all adds up. I mean, if you start discussing life and death, then most solutions aren't pretty; and discussing only 'pure' solutions means not saving (= murdering) lives - such as the lives of people that could have been saved by some sheep.
How about we don't murder any sheep or children, and instead spend the money that we would've spent on sheep torture on better fences, signs, education and mine clearance? While you're at it, you can lobby your elected representative to spend more than the currently pitiful amount that they do on mine clearance.
ps. A million-to-one? You have a very interesting idea of "ethics". Also of "definitely".
Sheep are scared stomachs on legs. When they see their daily source of food and nothing suspicious (such as a stranger), they reach. My grandfather could also call them and they would reach even with a stranger present.
Similarly, training locals in demining with donated basic equipment is also good, in the sense that these new experts can sell their services locally. The down side is that these brave people take extra risk.
Both the livestock and local training methods were used in Bosnia, which had a serious landmine problem recently.
i.e. you drive a stake in the ground on either side of the field (if that's even possible) and just create a lane where each of these rollers roll through.
It won't work in every single landmine, but it can provide a 'grid' within which you know what has been cleared and what has not.
May be harder with heavy / long lines / wind, but come on - we can build hanging bridges over very high canyons without anyone levitating to the other side first ;)
Mine rollers and mine flails like this have been tried and tested since WWI but none have proven completely effective in finding and clearing mines. This is partially because they only work well on totally flat terrain and rapidly lose effectiveness in rougher terrain where a large number of mines are typically buried.
In practice, solutions like this achieve only 50-60% effectiveness at clearing minefields, which makes them useless for civilian demining which demands a 99% clearance rate.
For that reason there's been a ton of research in this area which has resulted in better demining vehicles and interesting new techniques such as using sniffer dogs or rats to detect the explosives inside landmines. This is especially useful for the numerous plastic-shelled landmines that resist standard detection methods.
Using a layered approach with these techniques, civilian demining organisations like Mechem[0] (which pioneered the use of sniffer dogs) are now able to achieve a high enough clearance rate to make areas safe, though the work is expensive and time-consuming. If you support this sort of work, donating money to demining NGOs would be better than funding yet another ineffective mine roller.
[0]http://www.mechemdemining.com/
That's not a rhetorical question. I'm genuinely interested.
This method won't make the land usable safely, so the reality would see a bunch of cheap partial clearances with no real benefit - the areas will still be no go zones.
1. Is it really useful at all to only clear half the landmines? I would assume the area is still too dangerous to be useful even with only 40% of the landmines remaining.
2. Given #1, it would follow that additional (expensive) land mine clearing technologies would always be needed in addition. I don't know how these work or are billed out, but if the cost is per square mile or some similar area-based unit (instead of say, per mine), then clearing out 60% of the landmines first would not affect the supplemental cost. You've still got to cover the same area to catch the other 40%.
Anyway, you've had my disclaimer at the top, and I could be totally wrong about this :) Just some thoughts.
I'd be extremely surprised to find that a commercial clearing of mines for an area with 40 mines was the same cost as clearing that same area with 400 mines. Equally that clearing a double-sized area with 40 mines was the same cost.
Either way you'd reduce costs - reducing search area or reducing removal+disposals needed. [fwiw _djo_ intimated that the increased cost to the process of debris removed any utility such devices might have but also suggested that similar devices are in current commercial use].
With a sandy area I'm wondering how well a dragged scarifier like implement would fair.
And as another poster (codebeaker) pointed out, it's also far worse for demining teams to have random shrapnel spread over a minefield, as it makes the job of identifying mines much more difficult. That's one of the reasons why demining teams use systems that are less likely to break and leave pieces on the minefield.
So it's unlikely that this product will save any money.
My guess is that demining NGOs will be using something like this soon enough if it proves effective, though, so any donation dollar really is best sent there for maximum effectiveness.
The problem with communities doing it themselves is it'll give them a false sense of security about the safety of known minefields. This could potentially increase the number of mine-related injuries because far more people will attempt to live in or traverse known minefields which they wrongly believe to have been cleared yet which still contain roughly half of the original mines.
This concept is actually worse than useless because of the debris it scatters across a minefield and the false confidence it provides. While I applaud the designer's attempt to create something useful, his lack of domain knowledge has resulted in a design that doesn't solve the problem and may make things worse.
If I come across as bit hyperbolic, it's because I strongly believe that this is a horrible and underreported problem which demands both carefully-considered and effective solutions and acknowledgement of the people whose hard work and sacrifices have made so many places safe again.
It seems silly to laud an ineffective design while there are people out there every day clearing landmines with solutions that work.
You're saying something akin to: 'tyres are useless because cars also need engines; people are already effectively using tyres; often tyres don't work'.
>It seems silly to laud an ineffective design while there are people out there every day clearing landmines with solutions that work. //
Solutions that according to you - strongly implied eve within the parent comment - include rollers.
So
"Mine rollers are useless [for that purpose] and so should never be used to clear mines under any circumstances."
Yes or no?
Mine rollers can be effective for civilian demining, but only in the right type of terrain with the right type of vegetation and when used in concert with other demining techniques. I didn't just imply that it my posts, I stated it outright when asked when and where mine rollers could be effective.
But when used independently, as the Mine Kafon's video shows, mine rollers are useless for civilian demining and, because of the false sense of security they provide, could even be harmful. A minefield that people believe to be cleared, but isn't, is far more dangerous than a minefield that's demarcated and which most people know to avoid.
So I'd amend your provided phrase and say:
"Mine rollers used independently are useless for civilian demining and should not be seen as civilian mine clearance solutions."
I'm not sure how much clearer I could be on this.
"Mine rollers can be effective for civilian demining".
So, hang on maybe they're not useless as _djo_ was contending. But wait, why is it they can be effective we ask, oh right it's because they
"are useless for civilian demining".
The context of those phrases, which you removed, made it clear enough that the first phrase referred to mine rolling as part of a system of complementary techniques while the second referred to it as a technique used on its own.
By intentionally removing that context so as to attempt to prove the phrases contradictory, you've proven only that you're more interesting in trolling than in debating this seriously. So I see no point in continuing this conversation.
The point was that you started by saying the thing was useless and persisted to try and defend that point despite acceding that similar devices are used and useful.
There was nothing to debate, I simply called you on over-exaggeration and you refuse to accept it is. Fair enough you clearly know more about the domain in question. But the information you've given is the counterpoint to your initial claim.
The reason I quoted without the full context was because that indicated your initial argument. The extended quote was how you adapted your argument against the claim that you were wrongly portraying the true position.
I was an SFOR-soldier in Yugoslavia and I saw several crippled kids sitting next to a football field watching other children play in tears while others were reading superman comics which informed them about not touching mines and how they look like. They deserve hope and not a joke ..
Provided the weight came mostly from the core and the core never itself got blown up, I'd saw it can clear more than a couple of land mines for the money. Perhaps, a land mine per plastic plate, and if an NGO is paying for Afghani wages, you can roll that baby all day.
Bonus: it creates a relatively "safe" path in its footstep.
Does it matter if you blow some mines, if you don't make a single square feet of guarantee mine-free soil?
The haphazard nature of the exploration is the big issue and using this would most likely give people a false sense of security. This is an interesting piece of lo-tech design created with the very best of intentions, but it is essentially useless and may even be harmful.
A patch of land is unusable while mined; useful when cleared, but if it's 50% cleared (or 80% cleared) then it is murderous, because then it gets used, resulting in dead and crippled locals.
A field that you must use to avoid starving has 100 mines in it, you can remove [on average] 55 of those mines using a locally made and affordable device. The field will be used anyway. Would you deploy the device?
You're leading a demining team. A village have asked you to clear a field that they rely on for crops (other areas are too rocky or haven't enough good soil). You can send out some locally produced cheap demining devices first that will remove about 50% of the mines reducing the risk to your team. Would you deploy the device?
"is not useful" you say?
Your position appears to be a counterpoint to a statement that these devices will clear all mines effectively from all areas in all conditions. I've not seen anyone make such a claim. Your position is only tenable when a mined area is not needed by the locals.
The devices, and the beautifully made film, have another purpose which is far from useless too - attracting attention and thence funding for demining projects.
The argument for partial demining leading to increased harm requires that land that is mined remain completely unused otherwise or be completely demined. As soon as there is any use then reducing the number of mines statistically [and yes maybe simplistically] will be better - which would you rather, walk across a field of 200 mines or 400 mines?
Professionally demined areas are only cleared to a level of about 99.6% according to the thread so the argument for increased risk of harm still works in that case - more risk from mines when walking in a 'completely' demined area than in an area in which mines have never been placed.
Congratulations, you have imagined an instance where the this device can be used. The constraints you have imposed are: that the land comprising the minefield must be used and will be used regardless of whether the mines are cleared; that there is no predetermined upper bound of what constitutes acceptable civilian casualties; that the cost of subsequent de-mining is immaterial (e.g., when the local population decides that maybe 50% wasn't good enough after all); that the terrain is suitably flat for the Kafon; and that the extant vegetation will not interfere with it.
Now, does the scenario you imagined actually exist in the real world? Or is this purely an exercise in mental masturbation? I would suggest that the scenario you imagine is not one that is common in the real world. Afghanistan is not the first place where mines have been used, and yet there don't seem to be many farmers in Vietnam or Cambodia re-purposing known, uncleared minefields. Almost all civilian mine casualties seem to be the result of people using areas where they thought there were no mines. My take-away from _djo_'s comments is that the Mine Kafon primary effect will be to make it very cheap and very easy to create large areas where people believe there are "no" mines, and that is a problem.
In engineering it is often true that a general solution is preferred to one which only works in special cases. The Mine Kafon seems to be, when viewed in the most charitable light, a very special-purpose solution which adds no value above the general solution. Viewed in a more balanced way, the Mine Kafon is a solution in search of a problem[1]. User _djo_ seems to be saying that, there are already tools which do what the Mine Kafon does, and they minimize not just the economic cost to the de-mining organization but also the human cost to the local population.
[1] Do you recall the article several days ago by Paul Graham about how to find good start-up ideas? One thing I took away from that discussion is that you should try to find solutions to existing problems instead of looking for problems that fit your existing solution.
[0] http://www.ted.com/talks/bart_weetjens_how_i_taught_rats_to_...
"The Mine Kafon has just been tested by the Dutch Explosive Ordnance Disposal Unit, which has concluded that it is not suitable for mine clearance (which requires a more systematic approach), but at $50 each could be used as a cheap and safe way to identify dangerous areas that need demining."
[1]http://australia.icbl.org/index_htm_files/May.pdf
If you manage to clear a field such that there is a single mine in it, it will likely become safer to march through than avoid - there are plenty of other things that can kill you, and they must be triaged in war time.
However, if you take that same field after the war is over and put a farmer on it, it's basically a matter of time until he steps on it. His only chance is that an animal blows it up, or the mine is defective.
Therefore, if it's not suitable for military clearing, it is even less suitable for civilian purposes (except for figuring out what needs to be cleared, as they say).
[Added] I’m talking from a US perspective, but it seems plausible that most developed country militaries spend on technology when in time of peace.
All this discussion is because of an over-enthusiastic headline, not what the tool is actually intended for.
Well, this tech is not for the authorities. Its for children to use to help clear the danger away from their villages. Its a tech that anyone can apply, not just those with access industrialized manufacturing facilities and the 'mine-clearing economy'.
99% clearance rate? I think most villages would be happy with being able to clear a safe path for their kids to walk on, at any rate. This cheap, useful, easy to apply technology allows that to occur. Make 100 of them, let them loose for a few weeks, and suddenly 99% seems a lot more doable with this method than any other - especially for the resource-lacking Afghani villages who are left with the war liability that the imperial, authoritarian states, have given them ..
When you demine an area, you signal that it's safe for the population to move in, which means that if you haven't actually cleared 99% of the mines you're increasing the possibility of a person stepping on a mine. That's the reason the UN demands such a high clearance rate before an area can be declared safe for humans to live in.
http://en.wikipedia.org/wiki/Demining#Plants
Is that an actual official classification? It seems a bit low. "There were 1000 mines here, and we found 990 of them. Good enough, I say. Move in, and bring your kids!"
For instance, Mechem's sniffer dogs have displayed a 90% success rate on target filters, so statistically having three dogs per team perform cross-checking paths on the field should produce the 99.6% confidence level for the type of mines the dogs can detect.
[0]http://www.un.org/Depts/dha/mct/rptwg2.htm
I know nothing at all about this area, but presumably this only works if the 10% failure rate is random rather than based on some quality of the undetected mines?
But of course that isn't true in reality. There will be properties of the mines (burial depth, ground cover, degree of leakage of sniffable explosive material) that affect its chance of being found by all dogs searching for a mine.
If you wanted the parent poster's statistical chance, you'd want three 90% ways of detecting mines that are truly independent. Perhaps a dog, and infrared imaging sensitive enough to detect the explosive material (I have no idea if that's possible), and this article's mechanism.
And there are even mines specifically designed to be resistant to mine flails https://en.wikipedia.org/wiki/MAT/6_mine
Still, I agree with your first point.
http://edition.cnn.com/2010/WORLD/africa/09/07/herorats.dete...
http://minekafon.blogspot.nl
http://news.bbc.co.uk/1/hi/3437019.stm
It wasn't long until the locals just farmed snakes to sell to the British.
Plus, retrieving landmines would be a pointlessly dangerous task, when all you really need to do is blow them up when nobody is around.
Oh, and another one that has been popping up a lot around here lately is "vanishingly" (I've noticed Patio11 likes it a lot!).
With that said, props to the designer for working on a solution to a difficult problem, but perhaps more domain knowledge would have lead him to a more suitable solution
That would seem cheaper here too, and the still-alive goats can make milk.
(Many mines are also designed to be triggered under a certain weight, so goats can walk freely on them and not get hurt)
One interesting method I read about is seeding the area (from the air) with a certain flower. When it blooms, the contents of the ground affect the color, so on an aerial photo you could locate the mined areas.
Dedicated and skilled engineers best of course.
My understanding is that manual, low-tech demining is not actually that dangerous. It requires some training and caution, but generally speaking anti-personnel mines aren't likely to do you much damage if approached carefully from the right angle.
Whereas the approach of getting machines to jump up and down randomly is expensive, likely to lead to a false sense of security and may even cause mines to get pushed into positions that make them more dangerous.
You could get a lot of Afghan workers enthusiastically doing low-tech demining for not that much money. Every now and again one of them would lose a finger or two but maybe that's not such a bad outcome considering the alternatives. People tend to get much more excited about expensive, higher tech solutions because they're sexier and the people who develop them (often the same people that made the mines in the first place) are better at selling them.
Or you could just create a fleet of cheap ground stomping robots with armored feet designed to deflect the blast wave that are powered by cable and remotely controlled.
In practice the best way to clear most mines (to a military standard, which is far lower than civilian) seems to be explosively generated overpressure -- they launch a bunch of explosives, usually a linear charge, in the direction of travel, and it drops and blows up, clearing a path.
Using trained rats to clear mines (cheaper than dogs) seems to be the best upcoming way to meet the civilian standard. It's labor intensive, too, but most countries with landmines have relatively cheap labor, and the training to be a rat operator isn't as difficult as to be a full EOD tech. It's definitely one of the charities worth supporting.
as others have pointed out:
- your goal is to clear an area, hence you need a systematic approach to be able to deem a strip of land safe. random paths do not help
- this identifies and detonates in one go. but if the mine does not get triggered, the identifier is lost. these are two separate tasks. merging them would only work with 100% detonation rate.
- this does not work against other types of explosives. anti-vehicle mines, unexploded ordnance, cluster ammo,...
the upside is definitely the low cost. the low effectiveness rate and unreliable pathing makes it lossy though. which with mines is not acceptable.
http://www.youtube.com/watch?v=dkVW09X-h8I
http://www.travbuddy.com/Kulyab-travel-guide-1325354/photos#... http://serious.tumblr.com/post/39645197/beware-of-landmines-...
Could it be that the sign is in Tajik and not Russian? (I speak neither of those languages, so I wouldn't know…)