5 ms·
Help me explain farmer logic to tech folk
I am trying to get a conversation going between technical people doing robotics and small farmers like me.
I am organizing small round table talks to get conversation going.
Here is one thing that I am struggling to explain. Many farm robots are designed for large commodity crop farms and it seems like there isn't a lot of opportunity in small speciality food crop sector.
I think they are wrong. Why?
1. Small farms need more help. Large commodity farms are already mechanized.
2. Commodity farms are several thousands of acres but $ per acre is meager.opp is true for food producing small acreage farms.
3. Margins are thin in small acreage farms because of labour. If we close the gap, there is room for optimization. The opposite is true for large commodity farms.
And yet, I don't understand why tech and Silicon Valley finds large commodity farms more viable than small food producing farms?
We have an area that has room for improvement and to increase profit margin.
The VC investment enthusiasm trend in large heavy autonomous systems seems counter intuitive to me.
You can always scale up in farming. You can't miniaturize and halve/reduce the cost of giguantic robotic systems by scaling DOWN.
I am struggling to get my point across. How can I put it across in concise manner and short sentences?
- allears 9y agoPerhaps you need to look at the tech company perspective. They have to make a profit on the development of expensive hardware and software, and they may not sell many units -- it's not exactly a mass market. So of course it makes sense for them to go after deep-pocketed corporations, rather than small farmers. Rather than expect developers to think like a small farmer, try framing your pitch to developers in terms of how they can fund their development, and how they can then scale up sales volume to justify their investment. Or, try organizing the farmers to form some kind of entity that has numbers and money behind it, to attract the attention of developers.
- jelliclesfarm 9y agoOk. I hear you. I will work on it and try it again here.
- _dps 9y agoActually this is a topic very near to my interests. You mention it in farming (and that interests me, but the extent of my farming knowledge is growing a small garden of tomato plants) ... but I think it's a more general trend that technology tends to be developed with an eye toward massive scale, with small-scale efficiency as an afterthought. Could you give some examples of opportunities you see in developing technology for farming at the small scale? In particular, examples where the technology designed for the large scale farms wouldn't be appropriate?
- jelliclesfarm 9y ago1. Large farms have already reached peak mechanization. 2. Robotics is for automation, not mechanization. 3. For large farms, drone technology, precision application of pesticides/herbicides/fertilizers are areas of opportunity. Smaller farms can never hope to achieve the economies of scale that the big farms can easily grasp. Smaller farms need modular solutions. The knowledge to grow diverse market crops in a small acreage is an entirely diff skill set. Small farms don't need 'intelligent' systems. We need automation to reduce labour costs. Robotic solutions for small farms may reduce our labour costs and we could generate higher profits. AI and robots for larger farms might make the operations more expensive and not increase profit margins. AI and robotic systems for large operations will be integrated with existing big tractor companies. Maybe licensed. Which opens another can of worms. Farmers will never own the proprietary software or algorithms. Farmers generally don't like to use anything that they can't understand.(already happening with John Deere and DMCA). This is mostly because harvest/planting/feeding windows are seasonal and varied every year. Because of perishable nature of our products, even a slight error in timing will ruin that year's income. And you can't wait for someone to debug your tractor or trust that it will be fixed. One area where large farms are suited for both intelligent systems(AI..suitable for large farms) and automated systems(robotics..suitable for smaller acreage farms) are those working in vineyards. It's almost like a warehouse environment and it's a premium product. It can handle it.
- hanniabu 9y agoI'm not the parent, but I believe you failed to answer their question. He understands the issuend. He's looking for direction on a specific example for what exactly you would like automated and why the solutions for large scale farms wouldn't work in this scenerio.
- rpeden 9y agoBased on your post and your messages, it looks like you're saying that robots and automation could help small farms most by reducing labor costs. So it sounds to me like the kind of automation that's needed will be very task specific. You might find it easier to connect with the the folk by describing some of the specific laborious tasks you'd like to automate. That will get their tech brains thinking about specific solutions to your specific problems. Perhaps start by picking 2-5 tasks that are repetitive and take up a lot of your time or require you to hire lots of people. Some of them might end up being quick wins that wouldn't be too hard/expensive to automate. Providing the tech folks with a list to choose from will help them focus on the items that they know they can solve in a reasonable amount of time.
- jelliclesfarm 9y agoI have about seven repetitive manual tasks that would be easily accomplished by automation. I will refine it further and translate with maybe visual snippets. I was thinking of maybe showing it as graphic novel strip because I feel like a lot of times farmer's and engineers speak a diff 'mental' language. But taking a ripper shank to compacted ground illustrated with a 'thwack' inside a serrated thought bubble can explain more than if I explain in English. The knives and rippers and v-shanks we sometimes use as implements is very difficult to explain in words. A lot of times, farmers just tinker together stuff in their shop to suit their particular soil condition or that year's weather challenge.
- jelliclesfarm 9y agoThank you.
- sova 9y agoApproach it from the numbers: You need 1 big farm buying one $2 million dollar unit, Or you need 100 small farms each buying one $20,000 unit. If you can make a compelling case that there are many such $20,000 units (for example) that could be researched and invested in, and you have some strong ideas on devices and their network interplay, you could be onto something very strong.
- jelliclesfarm 9y agoThat's the cost of a small tractor already. It takes additional implements and man hours to operate it. If a 20k base module can exist like ..say..an iPhone and there is a library of tasks that can downloaded..like say a kindle book. I think we can see cheap tech. Let's say: clay soil + Swiss chard + zone 8A + September-December. We know enough to come up with a crop plan. As it's a shared library, with time, we can get whatever disease or pests come to that zone at that time of the year. What nutrients need to be added. Soil probes in the base module can figure out what's deficient and we can know what to supplement the existing soil with..connected with gps, weather data and past years pests issues/crop plans, you have a reasonably proficient farm assistant that can go into a mapped field and recognize the chard row..and scan for bugs or do a soil probe every 6 weeks or check if it's ready for harvest and predict how much will be ready for market every week etc. it's possible.
- johnwaynedoe 9y agoI think examples of jobs that have a high potential to be automated could help. Just to get the creative ball rolling. Or just presenting specific problems. The tech crowd loves to dive deep into an issue and try and fix it. But I think part of the problem is, they don't have enough understanding or knowledge to know what could be automated and where the opportunities are. For example, I grew up in a rural community, but was not directly involved in running a farm. I am interested in what you are saying, but I have trouble thinking of what needs automated. I want to put my brain too it, I just don't exactly know what I am putting it to. Hopefully that makes sense.
- jelliclesfarm 9y agoYes, I hear you! Because I feel the same way when I sit down and try to explain with words not the design of my robot but its function. I think I tried with 5 people who fabricate custom machinery. The only ones who understood what I was trying to say are those who have had some field knowledge or some farm background. No task can be completely automated without some kind of code making decisions. That's why large automation machinery meant for commodity farms are like blind farm labourers. They can't make decisions. Like pruning autonomous tractors or something that mows down or discs 500 acres. But completely diff and won't work when it comes to a bot harvesting strawberries. Some do..if they have a vision system or has some kind of intelligence system making decisions. And that's why it will be several millions of dollars. Nothing is as efficient as a human making decisions in the field. But our jobs can be made easier 1. if the data we need to make decisions is easily available. 2. If there is less manual labour. Smaller farms are great for robots. If there is disease in a 500 acre corn field, how would you figure it out? But if it's 100 five acre blocks. And we have a swarm of 10 bots that can handle five acres at a time, then they can work as a swarm and also collect information to compare data with other blocks. My point is that if we create intelligent systems AND autonomous solutions for small blocks/farms, they are scalable and can be replicated. And diff data sets can be created. But if something is created which functions only with a minimum lay of 500 acres, it's not only expensive but limited in its application. Also..it would be easier to customize and build on top of existing applications because each year farm conditions change and small farms are more diverse. I don't know if this kind of thinking is right or if it makes sense. But that's my reasoning for why one needs small acreage robots rather than large commodity crops farm bots. Thank you! I would like to know if I am able to be articulate. I still can't be precise and say in short sentences which seems to be a desirable thing for tech people. But it's tough for me.