 I am Athena and I'm here with Lisa and we are interviewing Craig Piggott. He is the CEO and founder of Halter and we're going to be talking about how artificial intelligence is evolving the farming industry. And then after that we're going to have your questions and comments from you, the community of tomorrow. So stay tuned. Tomorrow's Science Discovery 1.09 begins right now. I'm super excited for this interview today because we have the CEO of Halter, Craig Piggott. He started the company back in 2016 and actually won last year, the company won the New Zealand Innovation Awards and Craig is a finalist this year for the New Zealand High Tech Young Achiever Awards. So Craig, thank you so much for taking time out of your Sunday, early morning Sunday in New Zealand to be with us today. We really appreciate it. Not a problem at all. Thank you for having me. Awesome. So the first question I actually wanted to ask you today is what brought you to actually starting this business and starting Halter? Yeah, so well, I guess, firstly, the important thing to note is I grew up on a farm. So everything I did when I was younger, when I was going to school, that kind of stuff was evolved around farming. And it's everything my family had done, my parents had done. So that's probably the important part. And then I went off and studied mechanical engineering here in New Zealand up at Auckland University. And because of that, actually, I found myself working at a local company here while actually he'd put it over over in the US, but called Rocket Lab. And as a mechanical engineer, this was like my dream job, working for for a rocket company. But pretty quickly, it become clear that Peter Beck, who started Rocket Lab, was sitting there in New Zealand, creating this entire new industry, like the space industry in New Zealand just didn't exist. And Pete said just just creating it and the industry we're most known for the dairy industry just has so many fundamental problems. And really, I was sitting there being like, hey, well, I understand this industry. I grew up in it. I feel like there's a huge opportunity here. So ultimately, that was that was really the founding moment. I'd seen, you know, a high performing team operate at Rocket Lab and they go through equity raises and seen that whole, I guess, journey. So I thought, why couldn't we do the same thing with with the different industry? And that's really the founding moment. What about the dairy industry? Where are their problems? So where are the problems that you were discussing earlier? Yeah, so the biggest fundamental problem at the moment in the dairy industry is revolts like is based on labor. So just if you go talk to really any farmer, they're doing 80 to 100 hour weeks. They can't find good staff. No one really wants to enter that industry anymore. The what used to drive a lot of young people to enter, say, as like a farm assistant or a herd manager or any kind of entry level job on the dairy industry was the pathway from that to owning a farm. So those down age, that pathway is just so, so long and seemingly impossible that no one really comes in the bottom with the vision that they're going to own a farm. So because of that, there's this match of shortage of labor and results in things like sustainable like sustainability, sustainable farming, for instance, and be able to fence waterways. You just don't really have the time to be able to do that kind of stuff when you're literally struggling just to just to run all your usual activities. Wow. And your device is going to take away that need for labor. And in moving cows to different places in paddocks, we've actually got a really cool animation that helps to show what the cows are actually doing. Can we talk us through a little bit about the device or the whole device and how that that moves cows around? Yeah, so kind of the fundamental concept behind like everything we do is we build this device that gives feedback to a cow. So it's based on Pavlovian theory. So if you've ever heard of like Pavlov's dogs, where if you ring a bell and the dog drills, effectively, those two things aren't associated at all. Like there's no reason that when you ring about the dog should drill, but you condition the dog to respond to the bell. So it's a very, very similar concept to that. This device conditions the cow to respond to audio and vibrational cues. And then after that, we're able to literally move a cow left and right and forward and back. So it's an unbelievably powerful like re-association. But if you want to think about it from a fundamental level, it's really no different to say a dog barking at a cow. We're literally just training cows to take what is usually a visual cue, such as a fence or a gateway and associate that with an audible one. And this whole process only takes about like some of the smarter cows that can be two to three hours. And some of the other ones would, you know, might take as long as six hours. But fundamentally, it's still very, very quick. Wow. So how is that actually giving you guys the data and the information for whether a cow is eaten enough or needs something? What are the follow up procedures if, say, a cow didn't eat enough? Yeah, well, this is a really interesting thing. And it's something that we got, I guess, more and more clarity on the longer and the longer we've been in field, I guess. But you put in a sensor and a device, just take an IMU, for instance, that can tell you things like whether a cow is laying on its side sick or how much, how long it's spent with its head on the ground, eating pasture or really any kind of classic machine learning technique is nothing like that new. People have been using sensors to detect status and things like walking, running like your phone does this with you. And that was we knew in this for ages. Now, the problem is that typically that hasn't provided enough value to a farmer that you could justify putting the device on a cow. So where we found ourselves, though, was because we can be shift a cow that has enormous benefit that removes almost like half the labor on a dairy farm, removes the need for every single fence, every single dog and bike. And you literally free up so much of the farmer's time that he can spend doing other things or at least not working 100 hours in a week. So you've got this massive benefit, which gets the device on a cow. And then once it's there, you can do some really cool things. So you touched on there if a cow hasn't eaten enough, for instance. So how that can work is you've got 400 cows saying a herd, they go into a paddock and all the aggressive and the bigger ones eat first. And once I've reached their limit, we actually pull them back off the fresh break. So off the best grass. And then we allow the kind of the weaker or the slower cows to still be in there accessing the longer pasture. And if they take some longer to eat, that doesn't matter. But at the end of the day, they all eat the same amount. But at the moment, it's a little bit like a hierarchical race where the stronger ones eat more and get stronger. And the weaker ones just end up going hungry a lot of the time. Wow. So you mentioned, you know, you're combining that data from being able to move the cows, but also the sensors that are telling you, you know, is the cows head down near the ground. So how much testing have you done on this whole technology? Like what kind of stage of development is the device actually at? Yeah, so the short answer is we're going through production at the moment. So we're literally just getting everything ready to scale through mass production. So we're working with contract manufacturers in China and just trying to make sure that we're ready to roll this out because the average size of the farm is around 400, 420 cows. So you can see reasonably quickly, we need to be able to produce, you know, 50,000 of these a month because it doesn't take too many farms to really push that demand up. So that's been our biggest focus at the moment. We've been in field testing on our own development farm for just over two years now. So we have we're headquartered in Auckland, New Zealand. And then literally a couple of hours away, we have a test farm or development farm and that is the biggest asset for us as a company. It means we can very, very quickly take our technology, take the algorithms and the devices on to farm and actually put them through like real world testing with, you know, with farmers and with actual cow sheds and all the all the gear that you want. And you can do as much like whiteboard theory as you feel like. But it's not truly until you get in the field and start to understand how how a cow is going to respond to this, how you're going to build trust between the device and the cow, just like, I guess, a rider and a horse. Things that later are all very hard to do in theory. So the best practice is just to get in field and start trying to do it. Speaking of how this actually affects the cow, we have a question here from Space Mike and he actually asks, is this humane and do the cows experience pain from these signals? Yeah, so like first and foremost, we love our cows and this is like half the team here is from a farm. And if you talk to most, well, really any farmer, they're actually like super passionate about what they do in the land and the cow. So definitely doesn't hurt the cows. When you put a device on a cow, you all of a sudden have 24 seven monitoring of that cow. And right now, if you're a farmer and you have a thousand cows, it's really hard to keep track of what all them are doing. So we know if a cow has a sore foot seven weeks before a human can tell by eye. So things like that is just like absolutely game changing for the health of an animal. And this is because like cows are stoic, so they hide their injuries. So they will not learn until it is really, really bad to the point where they physically can't not learn. But something they'll do a lot earlier than that is they stop walking voluntarily. So you put a cow in a paddock and it will just stand there and sit down. Now, if you've got a thousand cows, you don't really know. But we know straight away. We're like, oh, this cow stopped eating, this cow stopped walking. She's actually showing small signs of a change in the gate, which you can pick up on an IMU. And then you can feed that information through to a farmer. And what it means is, realistically, when you like drink, this was a glass of milk as water, but if you had to drink your glass of milk in the morning, you can be like assured that it's come from a farming operation where all the cows are at a certain level of health and potentially happiness, but they're also not on sensitive environments and rivers or walking through low lying wetlands and things like that. So ultimately, this is like if you're a conscious consumer, this is a huge shift in the industry that needs to happen around animal welfare and trying to improve the standards of operation there. So to put it in short, the cows that we've been working with on farm, so some of those original, you know, the first, I think, herd of 30 that we started working with, they're the friendliest cows on the farm by a long way. How many have you worked with? Question. So our development farm is just under 300 cows. That's so we're going to be strategic about like what ages, breeds, dominance, we kind of cycle through every situation. This is where the AI machine learning aspect comes in, because every animal is so different that you need to understand like how sensitive that cow is to say sound and how agile they are. And if they're like slightly older and slower, you have to be more patient and really the only way this wouldn't be possible without like some form of machine learning because you just you couldn't physically categorize and code every different cow and how they behave. That leads incredibly well into this next question that you've kind of half answered it already, but from the chat room, honey's wall work from Twitch asks, does your device give us a profile of each cow? Can we know them on like a personality level to know if they're acting normally or if they're acting sick? 100 percent. So we already have kind of profiles for a cow because at this stage, we like switching devices a lot and, you know, we get a new version in from and from China and that will have maybe like a bit of a bit of battery life or what would be more durable. And we have to literally download, say, Cal 300's profile and put it into into the next color and then go from there. So like fundamentally, you could you could get to know a cow personally if there's the system set up to go right the way through the consumer. But in terms of things like sickness and how that is yet, it is very, very easy to monitor to monitor when cows are acting not normal as such. So they're reasonably predictable, predictable animals. Wow, there's a really good question here from the chat from Dada. He asks, does the system still work with high ambient noise environments? Yeah, well, this is quite interesting because when when we're doing this really early and when you the theory could work, when you that you could train, well, we tested pretty early and we could train a cow to to respond to to noise. But the concern was if you've got 400 cows on a paddock and they're all got like, you know, beeps and tones and they're all spinning circles and stuff, it's just going to be chaos. But what we actually found was because the collar sits right behind the ears of the cow, like literally probably two to three inches from the ears and the cows ears are massive that you actually can't even hear the sound. So they respond so well to the sound that before it gets to any volume that's audible for us. And I'm talking like if you were standing right next to it, you could probably within half a meter, you could hear it. But if you anything more than three to five meters from the animal, you just can't tell. So it's actually a very specific noise that the cows get very used to hearing and respond really well. So it doesn't really matter if these music playing from a house or people talking next door, it's it's quite a distinctive sound for the animal. Awesome. So you're based in New Zealand and New Zealand is famous for sheep. We actually have we have a comment in the chat room from Nigel off YouTube who asks, why just cows? Why not sheep also? New Zealand is famous for sheep. Yeah, well, this is a really good question because if you think about every animal and when you're trying to train any animal, what are the fundamental principles of training an animal effectively? And that's you have to be consistent. You have to be repeatable. You have to be patient. Now, these are all things which software are really, really good at, right? Like, that's exactly what software does and humans are really bad at. So you think about like when people try and train their dog to sit and it might serve and they'll they'll give it a tree and then it won't serve and you still give it a tree and say, like, whatever, and you do that maybe like once a week. And three months later, you hope that it sits. Now, fundamentally, every single animal in the world like should have some kind of like training or help with the decision making from like some software application. So no, I truly believe that even so your domestic dog could have a system that help train more so than just just a human. So sheep, that's one that comes up a lot, especially here in New Zealand. I'm not a sheep farmer. We haven't tested that. So I can't comment, but I believe so. I guess, yeah, you can look really through any animal, especially any animal that's that's held in like a paddock or with a fence or anything like that. What are the environmental impacts that you foresee this type of technology making? Well, the most obvious one in New Zealand here is we have currently a big thing with waterways. So having nitrogen leech into a waterway is like very bad for water quality. And the worst thing is when you say cows are actually in the water. So there's this big push at the moment to fence every single waterway in New Zealand. Now, there's like a lot of of kilometers of fence to do that. And it's often a cost that a farmer can't sustain. And it's also like a time like it takes a large amount of time to be able to do that. So the most obvious thing for us is when we always come back to us like, hey, we can keep every single cow that has a color on out of a river or out of any waterway. And that can be a seasonal waterway. So in winter, it could be wet and have like actual kind of surface water and a cow is in the summer. We could graze that because it's got grass in it. So this is kind of the concept we can do now. The other things that get quite interesting is just through the device, we can tell when a cow is urinating. So that's just through the machine learning model that that measures kind of motion and patterns. And so we know wherever there's been nitrogen managed or discharged into into the soil. So you can start to try and do some things. We was kind of looking through this stuff at the moment, but try and like even out the spread of nitrogen. So there's no highly concentrated areas that leech into the water table. So that's I guess one really good example around the one of the downsides of running like a pasture based dairy farm system as you have the nitrogen issue that we're trying to work on here. And I think if we can get that to a really good spot, that's going to be like unbelievably powerful and a real big asset for the future of of this like farming industry. Wow. I just love the whole thing about managing nitrogen, too. I didn't even think about that as an application. But you could even like train the cows to like go and lay nitrogen on different like areas of a farm and like you could have the like fertilization for free from the cows like close that cycle. That's so cool. But coming coming back to the the device itself, we've got a couple questions in our chat room here that kind of talk about the security of the of the device. So the first question comes from to wicked off the chat room, who says what guarantees can farmers have that someone couldn't just hack their herd for personal amusement or sabotage? And following on from that, Darknaze also asks, does the system use encryption? Yes. So from the from the start, we've been very, very aware of of the security security issue here. And I guess when you like these 1.5 billion cows in the world, so he's actually more cagey of cow in the world than there is people. And it's been pretty obvious that that is like down in the future. If you have kind of a large amount of cows under management, then you have a big security issue. And like that's been in grade from like our investors, from our board, from the team internally here. It's very, very serious issue that it's kind of like if you design it and if you're aware of it early, then it's a lot easier than coming back in a year's time and being like, oh, we need to try and add encryption keys to the payloads or whatever that is. But from the start, it's been it's been on the front of our minds and I'm pretty confident that the team in here has a good understanding of of how serious that is. Wow, there's another question actually regarding to the farmers by Hennies Vorwerp in the chat and they ask, how much work is it for the farmer to actually train these cows? And is it easy enough to not act as a deterrent? Yeah, so the the important thing to note is the device itself actually trains the animal. So that is like where from the start, we knew that if it literally meant the farmer spent more time on a keyboard or an app than he does in real life, then he's just going to walk outside and do exactly what he currently does. So that's been that's been like one of the important things from the start. And this is like this is pretty funny because as a start, you always run really lean, right? You always only hire like the critical people and you work on your biggest risks. Now, for us, one of the biggest risks has been like the user experience. Like how do you take a farmer who traditionally has a lot of tech that doesn't really like not that useful? He's sick of having or she's sick of having people turn up and say, you know, you have a three percent gain in X over three years and that you've got to do like all these things. You can learn this new system and it's just like too much and I don't have time, right? I'm time poor. I don't want to learn a new system, unless it's going to completely change what I do. So we've been working with farmers from like almost the very start. Like I guess I grew up on a farm. It was really easy just to pop around and visit like kind of 10, 20 of the neighbors. We hired for specifically for user experience from the start. And we spend probably like more time talking to farmers than really anybody else out there. It's either like head down in here, working away on different things, or you're sitting down having a coffee with a farmer talking to you know, what if we did this or have a play on like the app and give us some feedback? It's really, really interesting what you learn. They're actually amazing and smart people. It's just trying to get the time out of them to actually like make a difference and something like this. So how do you, you know, take a farmer that doesn't want to be overwhelmed by tech? How do you design that user interface? Is it as simple as, you know, like having an iPad and you see you're a farm and you just kind of draw boxes on where you want the cows to go and where you don't? Is it that simple? Well, this is so the simplest way you could do this, right? And this is maybe a little bit idealistic, but you could put devices on a cow and you could tell a farmer, all right, for the next week, just run your farm completely as you used to. So we sit here and we can see what tiny milks in the morning. So again, it's five a.m. And the cows with the shed, he milks at five a.m. You could do the same thing in the afternoon. OK, cows with the shed. He's milking at three thirty p.m. in the afternoon. He gives his cows, you know, July, he's giving his cows so many hectares or so many acres of pasture. And you can almost get to the point where after this calibration week, we just copy it and he doesn't even need an app. So he just wakes up at five o'clock and the cows are in the shed. And we do that because that's what he was doing. So that's like one extreme. Now, that's, I guess, not practical because different things, you know, if it's raining and it's two o'clock in the morning, the idea is that a farmer doesn't have to now get out of bed and hop on his bike and go pull those cows off the grass and put them on a pad to protect the paddock. He can now just roll over, pull out his phone and say, like, pull the cows off. So you're trying to make the interface like a very, very simple. So it is this balance between almost they don't have to do anything to, you know, they could have as much control really as they want as long as it doesn't go against kind of the fundamental principles of what we believe in. So we've got a few, like, I guess, underlying golden rules or guidelines, things like you can never walk a cow through a river. Even if a farmer wants to, we don't let you. So these these are the split between that. But fundamentally, it's a reasonably simple system to use. Yeah, it sounds very simple. So do you think this will ever become completely self automated where there will not be a need for a farmer? Well, this is like we live in this world of data, right? So literally, I don't know what the stats are, but I think it's like less than one percent of data is actually analyzed and used. And this is where we think we've got like a big advantage. And that's where effectively the first actually like or actual actuation on a farm, so we actually do things. And that means we can actually we can take in kind of all these insights, all these senses, now a great example is, say, like planet, planet labs and with their kind of constellation and satellites. And they can do optical, well, near infrared and optical imaging of of pastoral apathetic, and they can tell you where your grass is. And then we take a system like that and we feed that into HALM is what we call our software, like managed replacements. We feed that into HALM and all of a sudden as a farmer, you know, where your grass is on almost like a daily resolution level, we can take the best spots on the farm and put the cows there to eat the grass. And you don't have to do 400 cows on a paddock. You can do, you know, 10 cows in this area and 20 cows in this area. And you start to run a system that's like a lot more natural for the cows than the current intensive kind of farming methods. So it will become very interesting how, you know, how much influence the farmer wants to have over kind of just driving these decisions by data, because ultimately we are empowering a farmer. We're not making decisions for them. We do suggest a lot of things and a lot of these things are backed up by data. But if they say, you know what, I know my land better than everybody else. I know that paddock's wet, even though it's got a good grass, I can't put my cows in there. Then it's up to him or her that I think probably will evolve over time with more trust in the system and just more experience with more cows kind of under management. I think you guys are collecting a whole lot of data that, you know, maybe right now, maybe you guys are analyzing the data you're getting from the cows. But I think you're also, as well as your product, you're creating this huge database of cow behaviour data that people can use way in the future to kind of optimise the best way to raise these animals. So that's kind of cool as well. But coming back to making lives easier for the farmers, we have another question in the chat room here from Amy Giac on YouTube talking about do the cow halters have solar panels or do you need to go out and change the batteries? Is that another thing that farmers need to do? No, yeah, the colours are solar powered. It's effectively a non-negotiable on if you've got, say, 400 cows, just putting the colours on as an investment in time. Like, you know, if it only takes kind of five minutes each per cow, then you're still, you know, putting like a huge section of time and to get in a farm of money. So once these go on, they stay on for four to five years. They're a solid state. So typically nothing in the collar is a moving part or needs maintenance. There's been a huge part of the design for us is getting things like the power consumption so that it can run through like a foggy New Zealand winter and not go offline. So we do a lot of modelling around power consumption and use like a lot of low power sensors and low power, like microprocessors and things like that. So that the consumption is sustainable. So you have worked with over around 300 cows, you said, right? How many farms is that? Yes. So that's on. We've done all our testing on one like development block that the reason for that is that's been set up with a lot of extra infrastructure. So like, we have cameras all around the farm so we can run. I've even ran that farm from San Francisco. So we're like on a typical farm, you obviously don't have cameras and you don't need cameras. But for us, this allows us to test like new algorithms and or anything really remotely. And that's that's really important. So how many more farms do you foresee yourself expanding this technology to? Well, this is something we say quite a lot. And that's like we have very small market risk. Fundamentally, if we can get this to work at scale, if we can produce, you know, the number of devices we need to through a factory and we can assure the quality and the durability of these devices, it's a no brainer for a farmer. So and what we like to compare it to is like 200 years ago, you milked a cow by hand with a bucket and then someone came along and said, you know what, like we can do at least a better way to do this. And the milking machine was invented and then now nobody milks cows by hand. Well, effectively, nobody in the Western world milks cows by hand. And we're sitting probably in a similar spot where today you are the walk behind your cows, cows walk slower than humans. Or you're out there on a bike with a dog. You spend kind of hours a day following these animals around. And we're looking at a system that can do that for you. So I think fundamentally, the farmers that adopt a system like this will have such a strategic advantage that if you don't have it, you will not be able to farm. So we're probably sitting there in a spot that is there's no reason every cow, every pasture based dairy farming cow or beef cow or around New Zealand, Australia, South America, North America, Europe. Like, you know, kind of it's up to us like how fast we can roll out here. And we've always seen that from like, there's been huge interest from from industry with farmers that are soaking to help out. They offer up, you know, you can you can test on my cows, like come around whenever you want and I'll help you like develop the algorithms for feeding them and things of that. So it's very like, I'd say it's a very community based almost a community based project in a way. Wow. Do you have any restrictions in terms of like the kind of landscapes where you can implement this technology? Like if you had really mountainous areas or stuff like that, are they kind of hurdles for you guys to overcome? Well, we we tend to think that we're probably in one of the like harshest environments right now in terms of New Zealand. So like our weather systems are reasonably crazy. We have a lot of rain. We have short daylight hours in the winter. We are reasonably remote in terms of the rest of the world. We are at the bottom and on the other side. So I think that like if we can get the system to work and it has been working here really well, then there's no reason you know, everything should get easier as we start to look at say expanding through North America. But as far as like deep gullies and big ridges and things of that and theory, it's it's completely fine. It's just like how well that works over to call it like massive stations and things of that. So there's a little bit more test into it on there. But as engineering, that's you can solve those problems. It just takes time. I want to ask you a little bit about other types of applications that you think this technology would be able to go into. My own thought is I wonder if this can ever be used for crops for knowing how much sunlight or water is actually being being intake in certain areas. But there is a specific question that I really like in the chat from Hennies Warwick that asks, could a system like this be used on pigs to find early evidence of swine flu for disease management? Yeah, so the disease management is a super interesting part of this because a lot of those times you have signs that are very, very well known and predictable. And, you know, people pick up on these diseases. Typically, the first sign is some visual cue. Like you have a suspicion as a farmer that an animal is behaving weirdly and if you're a farmer who is picking up on this, it's it's pretty obvious if you can see it by eye, you can sense that through motion tracking or accelerometers. So the idea is that you could be like a lot more confident and detecting these things early and then work out how to manage them. So, especially in a lot of cases, say, so here in New Zealand, there's been like a kind of a topical thing around a disease put in bovis. And this is transferred through cows, touching other cows. So the only way you can go from one farm to another is when you put your cows in a boundary paddock and your neighbor's cars are in a boundary paddock and they touch, aside from literally selling the cow and leaving your farm and going to another. But so this is like where I guess we have a unique value proposition here because it's really easy to almost instantly say, you know what? Cows are allowed from two litres of boundary fence. That's just straight up rule. And then you've effectively isolated every single farm that either has this disease or therefore can't get the disease. Now, how this applies to other animals or anything like that. Like I think anything that has a characteristic that you can understand or you can define, like you can sense and then categorise an animal into. So that kind of stuff will begin will become quite interesting. And like aside from just getting on to on to say like a pig farm and actually testing that, my answer is like, I think so. Yeah, I think that would be really good for future technology. You're going to ask something, Lisa. Yeah, I just I'm so impressed by this whole technology and all the different use cases. But of course, for this to really take off, you know, the farmers have to have to sign up and start generating revenue for you guys. So this next question comes from Dada in the chat, who asks, is the investment in the system something that even smaller farms can afford? Or is it like a trade off to the purchase in that the farmers are only working like 40 hours per week now versus 100 hours per week? So what's what's the kind of feedback on that? Yeah, so we've kind of we've known from the start that like the fundamental thing here is it has to be affordable. If it doesn't increase the bottom line, ultimately, a lot of farmers that are a cash poor will say I can't afford the system. So the way that we're going to structure our business modeling, we're having these discussions just with farmers one on one. But as it's a monthly subscription, so you pay per cow per month. And effectively, what that means is the moment you first pay for like a monthly cost, you're receiving benefit. And it's immediate and tangible. So, you know, you put the colors on a cow on your herd one day. And the next day, you don't wake up at four thirty and the cows are at the shed when you when you go there. So this is like one of the strongest things that that I think attracts interest from the industry is that it's just so tangible, the benefits. You can literally see it work and see it happen. And it's not this blind faith. I mean, the whole trust me, this is good for your farm. So and then the way that increases revenue is there's a whole lot of very, very well known like farming techniques. These if you go into any like best practice farming site, it'll say, you know, you should feed your cows this much and you should back fence, which protects the grass when it's already been grazed and it's growing in a reasonably sensitive part of its growth cycle. You mean to fence that grass off from cows, re-eating it. But that kind of stuff just doesn't happen. So all farmers know the benefit to it. And they say, like, yeah, I wish I did instead of time. And when I see this happen just for them through the system, I they instantly get the benefits and how well that that will benefit like their situation. So that's what that means. We've received very, very strong signals from industry. It's also a really good business model for us. It means when we first put a system on a farm, that's the start of an ongoing relationship that we hold for the farmer. So it's not like we're just going to give them tech and it doesn't really work. And then we say, you know what, you've already paid for it. Good luck kind of thing. That's just not the approach we're trying to take at all. So right, those relationships are really important to build because it is continuous. It's a continuous industry. Well, that's so great. Well, Craig, where can everybody find out more information? So halter.co.nz. So that's our website at the moment. We're based in New Zealand because it's literally one of the best places in the world to develop technology like this. So like this is, we invented the electric fence, for instance, which is like a pioneering innovation for for farming around the world. And like we've got more cows here in New Zealand than we have people. And we had a questionnaire that's like, got so many sheep in New Zealand. Well, actually, we have so many cows. And so this is what we're based here. We, all our investment and support actually comes out of the United States. So we've got investors in San Francisco and in Chicago, things like our legal teams and all that are all over there. So it's kind of this like strategic step for us to be based here. So, yeah, halter.co.nz or NZ. Otherwise, yeah, Twitter, Facebook or the LinkedIn, all the usual social channels we're also on. So feel free to reach out really anybody. Awesome. Thank you so much, Craig. It's been a pleasure having you on. We wish you all the best with spinning up this device and getting it out there and making a real difference to farmers. And the way we as consumers, you know, consume dairy products and maybe even more in the future. But before we head to break, we do want to give a very big thank you to our patron supporters, specifically our escape velocity citizens. Thank you. Thank you. Thank you so much. These people contribute $10 per episode. They make the show happen. And honestly, we couldn't do this without you. We also have our orbital citizens who contribute $5 per episode. And again, thank you so much. I'm going to give you a bit of time here so you can find your name on the slate because every week we get more and more and more. And it just blows my mind how everyone supports us. So if you'd like to contribute to the show as of tomorrow, too, head on over to patreon.com slash tmro. And if you're not able to contribute financially, you can always hit subscribe and hit the little bells to your notified. Hit the like button. All of these things help us out. Share us on social media. Just thank you, everyone, for all of your support. We're going to head to a quick break now. And when we come back, we're going to have questions and comments from the community of tomorrow. Stay tuned. We'll be right back. We've always looked to the stars. They guide us. Give us comfort. Help us find our way. We see ourselves out there. When we look up, it inspires us. And we long for something we don't yet know. We yearn to go there. So we venture forth. We choose to go to the moon and this decade and do the other thing not because they are easy, but because they are hard. Because that goal will serve to organize a vision of this. And God evades here. The eagle has landed. That's one small step for man. One giant leap for mankind. The exploration of space will go ahead whether we join in it or not. Many think we stopped exploring, but we know our journey didn't end. We've only just begun. Ryan is functioning perfectly at this point. Come with us and explore tomorrow. And welcome back. So last week we spoke about origami underwater robots. It was really, really cool. We interviewed Brennan Phillips and Z. Ern Teo from Harvard University and University of Rhode Island. So we've got some really cool comments and questions that we're going to go over right now. So the first one comes off of YouTube from Louise Selsor, Jr. And they ask, they actually say, I love the experiment with the new format. The interview was dynamic and engaging. You two did a fantastic job. Keep experimenting. I kind of miss the news though. Maybe a shorter news section inspired by Mike's launch minute launch. Wait, Mike's minute launch. Or just read the post from the community about what recent science news they found more interesting. You can create a hashtag for that like hashtag best TMR science. I like that idea. I think that's pretty cool. What do you think about that doing a 60 second? I think if we're going to do science news again, it needs to be short and snappy. Because I feel like our science news would kind of drag on. But like this comment says, we're always experimenting with the show. And I don't know about you, but I really enjoy having the interview up front. These people are talking about really cool stuff. I like this. I'm having a lot of fun with this. I'm loving the double interview. The dynamic, yeah. It's really cool. It's a continuation kind of from space and space ends here. So I feel like it's a really good leeway into science. So I like it a lot. I think if we were to do something like Mike's Launch Minute, then it would have to be some type of science thing that would come now maybe. 60 second science? 60 second science, yeah, I don't know. But I like the hashtag. Hashtag best TMR science. Okay, yeah. But if you guys have any ideas about how we could take science news and make it snappy and just more engaging, I think is what I'm looking for and more energy, but still conveying the importance of that science. Let us know in the comments because just, yeah, what we were doing wasn't reaching that. And so rather than just continuing to do it because we've always done it and because it's what we do in space, we were like, no, you know what? If it's not hitting what we want, then we'll put it back in when we can get it right. Exactly. Awesome. Cool. Thanks for that comment, Louise. The next one also comes off of YouTube. It is from Zapfan, Zapfan. Zapfan, Zapfan! Okay, oh, I said it so corny. Okay, Zapfan, Zapfan. Yeah, you have to do the Zapfan, Zapfan thing. I think my Z's backwards, okay. They say, awesome robots. Could that sort of sample, wait, whoa, could that sort of sampling are be used on a satellite grabbing a chunk out of an asteroid? It could maybe be an alternative to the sampling mechanism on Hayabusa, wait, how do you say that? Hayabusa, I'm so sorry, guys. Or the PlanetVac being developed. Great comment. I think this one was, yeah, this was a really, really good comment when I saw this actually off of YouTube. And I'm not too sure about that being used on a satellite. I think it might, it's a really good idea because of asteroid mining and what's being done. And like, everybody's talking about doing like sample return missions, right? But like, how do you, even just attaching to an asteroid or a comet is really hard. Like, look at the Philae lander that Issa did to land on comet 67P, I should have known that. Charisma Nenchka, what's your bat pronunciation? Sorry, comet 67P. And Philae, they had these like arms that were supposed to like grab onto Philae when it hit the surface and it didn't work. And so it kept bouncing and then it eventually ended up in a permanently shadow crater and we lost the poor little lander that could. But if we had, you know, this tech, because I mean, it works underwater. So why wouldn't it work in space? I guess, just taking that device and optimizing it for use in space and then just using it to like, slowly like grab onto something. I think that'd be really cool. And then like, even if you used it like in orbit, you could have the object close around like maybe a piece of space junk. And that space junk is still gonna be like floating in the middle of that like round arm ball there. Yeah. And so. Like capsule. Incapsulating it, but not actually touching it as well. So I feel like it would just have some really cool applications in it. I'm like really excited for them to work on that. Yeah, we'll write a letter to them, pitch an idea, partnership with ESI or something like that. Hey, if anyone out there like works for a space company that would like to partner with these guys, you know, I don't know, I could make an introduction for you guys and I would love to see that working in space. Yeah. That would be really cool. Where's Vax? We need Vax to come back. Reaction spheres, grabbing arms spheres. It's just meant for him. Yep, exactly. Awesome. So our next comment comes from the community.tmr.tv from Follux. Oh, I just wanna get some context. It's a wall of text, I'm sorry, but it was a really great discussion relating to can we actually live on Mars? That was what the thread of the comment is from, so can humans live on Mars? Hey, you can read it. Yeah, but that's the context. Ooh, okay. So they say, according to Google's wild, unsourced speculations on Earth, we are seeing around 100 milligrams of dust inhalation per day. I have no idea if that is realistic for a controlled habitat, but we'll roll with it. If the dust is 1% per clit, sorry, that would be a daily inhalation dose of one milligrams, which is 1,000 micrograms per day. That's less than half of the 260, 2,640 microg per day dose, which begins to affect thyroid stimulating hormone levels. So at a glance, the numbers indicate we're probably going to be just fine as far as the thyroid function. As for Mars, because iodine is a competitor to perchlorate, I'm just gonna pronounce that word better. Everyone can make fun of me for this. Increased intake of perchlorate can be countered by increased intake of iodine. As long as a decent balance is maintained, thyroid function appears to be able to tolerate a fairly wide range of intakes of the two ions in people without thyroid issues. Thus, you could think of iodine's simplebation as the treatment to perchlorate poisoning, okay. I'm so sorry, I should have written that up, please. That was a torture. You did so well. Here's the big idea with this comment. What is the thing behind this? Here's the big idea with this comment. Everyone keeps talking about we can't go to Mars. There's too much perchlorate in the dust. The perchlorate, you're gonna breathe it in when you get back from a space walk and it's gonna kill your lungs and it's gonna kill your thyroid and then you're gonna die on Mars so we shouldn't go until we figure that out. But, like this comment says, and this comment in the community forums, if you don't in the community forums, go and check them out, they're amazing. This discussion was awesome because this person used scientific papers at every step of the way through this argument to back up what they were saying. And in a nutshell, it's basically, we're worried about perchlorates, but if we breathe them in, the problem with perchlorates is that they out-compete iodine in your thyroid. So your thyroid makes a bunch of hormones that help with growth and keeping you healthy and if your thyroid's not working, then it's a really bad disease and it really affects you with your metabolism and you have a really bad time. So you don't wanna hurt your thyroid and perchlorate can do that, which is why it's bad and which is why they complain about it from Mars. But this person is saying that if we just give the astronauts iodine supplements and we already have iodine supplements today on Earth, table salt is usually supplemented with iodine so that your thyroid stays healthy. If we just do that on Mars, but maybe a little bit more extra iodine, then we could treat, perchlorates wouldn't even be an issue because you're just supplementing with iodine so that the iodine goes to your thyroid instead of the perchlorate. So I think that's really cool. And this came from someone in our community that completely changed my mindset on whether perchlorates are gonna stop us from going home. Is there any research right now about this being done? Does anybody know? Not specifically from Mars spaceflight, but I feel like maybe, maybe this is a research direction that we should take because it seems like a really simple solution to something that has been tending to hold us back. Like, no, it's not the only thing holding us back from going to Mars, but it's been one of the points of like, we need to sort this issue out and maybe we already have. Just take a pill and... That's really, yeah, I'm really glad that you actually grabbed that comment as this torturer says it was to read it out loud, but that's really, oh man, that's really interesting. So I'm gonna look into that. There has to be research being done about that if somebody's thinking about it. There's this research being done by like, let's give, because perchlorates are also an issue on earth in like drinking water, like groundwater. If you live in a place where instead of like getting your drinking water from rain, maybe like drill really far down into the groundwater and sometimes perchlorates can contaminate that and that's like how we learned that perchlorates were bad for us, because people got really sick. So that's the kind of research that's going on, but like maybe we could take that research the next step and start giving higher doses of perchlorate and then like one group doesn't get iodine and one group does get iodine and see like, are they okay? We really, really cool. Really good potential. I love all these comments by the way about the moon dust right now. I just wanted to point that out because it ties back to science. So anyway, next week I'm really excited because we're going to have Sophia Nasser on and she's actually a PhD cosmology student and she is working on dark matter. She's over at the University of California, Irvine and she's working on the particle properties specifically of dark matter, which is a substance that makes up 25% of our universe. It's a really big mystery right now as far as the physical properties of dark matter. So I'm very excited to be doing this interview. So we're going to have her on next time and so make sure that you guys tune in. And you actually booked that guest, didn't you? Yes, yeah, we actually, we met on Twitter. So it was really cool and yeah, her research was so good. I know dark matter is such a great subject. I'm sure you guys are going to be really excited for that one. Yeah, I'm excited. All right, so before we wrap up for today, again, we want to give a very huge, amazingly large thank you to all our citizens of tomorrow that helped make this show happen. So our Skate Velocity citizens contribute $10 per episode. Thank you very much to all of you, but we also have our orbital citizens who contribute $5 per episode and a very big thank you to you as well. We couldn't do this without you. But also there's more. We have our suborbital citizens as well who contribute $0.50 per episode. And honestly, this list is so long. I feel like if you wanted to, you could totally be a troll and change your Patreon name to like Ben is always wrong. And then no one would ever like notice because of being in this like sea of names and you never get to work it out. And hopefully I've stalled enough that you can find your name. So there's that. We also have our ground support citizens with an even longer list of names. Thank you for contributing between $1 and $2.49 cents per episode. We really, really appreciate it. Again, we couldn't do this show without you. So thank you, thank you very much. And again, if you would like to contribute, head over to patreon.com slash T-M-R-O. Remember, you can always subscribe to us and hit the little notification bell so you don't miss when we have new episodes of Science. Speaking of new episodes, the next Science episode will be on October 6th, which will be next month. Until then, I think we're gonna go. So thank you so much for watching and see you next month. Thank you guys. Bye.