Nice project indeed, been following it for some time. I grew up on a farm, have agricultural and IT degrees and also worked in automation in farming, factories, cities and warehouses.
This project focusses on a technological solution to societal, economical and agronomical problems. You could try to make a cheap robot that does traditional farming like tillage, fertilizer and pesticide spraying, irrigation on a monoculture. Or you could design your farm with permaculture and foodforest principles which makes it like a self sustaining ecosystem that reduces a resource usage of water, labour, nutrients, pest control etc. Society leads people into bullshit desk jobs where they have to spend their free time in the gym to not become morbidly obese. I think society could benefit if people could enjoy at least part time doing manual labour growing food instead.
What kind of per-acre yield do you get with the methods you described above? Can they feed our population? Do the same methods work in coastal California and Indiana?
Well this does go back to traditional high yield techniques like terra preta which besides making superfertile soil has the added benefit of being a carbon capture and soil preservation solution or the three sisters system where you get yields from three crops on one field and even get nitrogen capture from legumes. What does high yield mean in the short term when you are destroying your most valuable resource. Even if you have high yield one dimension, how efficient is this when you have multiple inputs and outputs and you end up in a pareto efficient situation. The highest yielding farms do not necessarily have the best scale efficiency. Permaculture and food forest systems can have very high yields if designed correctly. California is losing thousands of acres of land to salinization because of bad irrigation practices. Permaculture means to improve and protect to soil and increase water retention with the help of carbon content in the soil. I'm not from the US so I would'nt know about Indiana.
There was a paper recently showing that small scale allotments have higher yield then conventional farming. The suggesting is that it was the high labour approach, including high crop diversity that increased the yield. It's this sort of thing that robots can do: run large farms like small scale allotments, with all the environmental benefits that would bring.
My parents had a highly computerized feeding system on our farm. In the summers I worked student jobs in highly technological green houses picking tomatoes and pig slaughter houses besides the work on the farm. I've grown some form of scepticism from experience of the high degree of automation in food production systems from a consumer or operator side. Do mind I work in automation but it needs to bring more value than just playing with toys. Permaculture systems are designed to reduce or optimize usage of all resources also labor. You set up an ecosystem where pests are kept under controle by other organisms instead of relying chemical or mechanical controls. This is even more puritanical than classic organic farming where you rely on natural pesticides. I am from Belgium where we have the belgium blue white cow which is very highly yielding in meat because of a genetic mutation but you are very dependent on vets for giving birth via caesarians because of the high degree of muscularity. There is a trend towards other races like limousin for meat production. In biological systems theory of constraint applies very much were you are very dependent on limiting factors like micronutrients, vitamins or in this case microchips. Humans and hamsters are both the only animals that can't produce vitamin C which makes fresh fruit a limiting factor for us or we get scurvy. The reliance on natural gas for nitrogen fertilizer via the born haber proces is already a handicap in a polarizing world. Imagine being reliant for operation of your food production system on a geopolitical hotbed like Taiwanese computerchips. I do think technology can also help with the design of robust food systems. I know the food forest institute in Belgium has a course on using drones for surveyance.
Are you able to recommend some resources to get started with permaculture. I'm also in 'tech' but I've recently been developing an interest in 'living off the land' and would like to have my feet in both areas going into the future.
I learned about permaculture from books in my native language (eg Louis De jager from food forest institute in Belgium). I'm no expert though. I did find docu like "The Biggest Little Farm" very inspirational.
Reinventing existing trust mechanisms but with an extra (computationally expensive) layer of things that can go wrong is the opposite of an improvement.
We have Yelp, Google reviews, Amazon reviews… how would something like that work on a blockchain/ledger and not be gamed just like the review systems we have now?
And how would we avoid this reputation blockchain from being populated by maliciously deceptive information, be it positive reputation promoted by the business through bribed or fake customers, or negative reputation promoted by unethical competitors?
I do think NFTs are a bubble now. By investors that think that one NFT is worth the same as the next one minted while it is just the exact pointe that they are not interchangeable. You don't get a piece unique with significant history attached. Mondriaan, Van Gogh and Monet were innovators now it is just a deep learning gan style trick. I do see the value of NFT in asset tracking of unique pieces to provide transparency, fairness and safety in supply chains.
Like another commenter said, you can do the asset tracking with digital signatures (owner signs that they've handed it off) you don't need blockchain, mining farms burning a small countries' worth of electricity, etc.
if you ever worked at the bottom of a supply chain you would understand that there are currently major issues with gatekeepers levying high margins (note I grew up on a industrial farm). If a farmer could sell his product under a smart contract that garantueed a percentage of the transaction costs flowed back to him and end consumers would have greater transparency into the origin of its product their would be a nett societal benefit. Minig farms blablabla, you can use other proofs for verification (proof of stake) or scale it with zero knowledge proofs
I don't know the specifics but most suppliers don't want to deal with the hassle of retail and marketing, so it makes sense for distribution / retail gatekeepers to charge a fee. And smart contracts aren't going to magically guarantee someone will buy your corn.
Moreover, ETH transactions often run in the $100+ range now, mind-bogglingly expensive and inefficient. Those fees pay miners to waste electricity, hoard silicon, and scorch the earth. Ethereum has been switching to proof of stake "any day now" since it's inception 6 years ago. And even if they actually pull it off I don't see why staking $20k of crypto to participate is somehow better or more fair than spinning up a $4/month cloud VM to manage a 1000x more efficient, fast, and effective supply chain database with digital signatures.
There's a lot of utopian handwavey BS around crypto, but it seems to be turning out like other magic utopian solutions in human history (gurus, communism, etc.). Wish people would divert this time and energy into things that actually benefit society.
Well I do know the specifics, I've worked on our farm. I've worked in and for processing plants and at food retailers. I did an engineering degree in agriculture with a minor in economics (plus some post grads in IT). We produced and sold our farm products at local farmers markets. I now work at the agricultural government agency of my local region. And I can tell you that the margins that those gatekeepers take are not proportional the service of marketing and distribution they provide. I'm not talking about magic. I'm talking about mathematics and logic in the form of proof systems that have the authority that would otherwise be relegated to central enforcers. There is distributed ledger technology that already uses proof of stake like cardano or polkadot. As for high transaction cost and scalability of pos you should look into zero knowledge proofs which allows to package a massive amount of transactions into one proof that hides a lot knowledge and still allows for checking transactions: https://starkware.co/
I defer to your domain expertise there, don't know much about the food industry and supply chain. Sounds like there are some monopolistic players & regulatory capture involved.
I don't think ZKP or blockchains are going to save you though. Say Cardano actually gets past their "Byron" ICO pump-and-dump phase and implements smart contracts, or Polkadot gets off the ground.
OK, so we have a fast, non-world-destroying blockchain computing platform and can freely write programs to exchange PolkaDots or Adas or whatever. So what? Now anyone and their brother can set up a futures exchange via smart contracts, and we can all trade PolkaDots for corn futures. Whose brother's exchange do we use? If a de-facto standard one arises, who enforces quality and resolves disputes if the produce isn't up to snuff? Blockchains tend to be immutable no-refund situations, and discourage the use of "oracles" from outside for human factors like the corn being moldy. Who regulates the exchange to prevent abuses on both sides? And couldn't MonopoliCorp still use its connections and infrastructure and reach to corner the market? You still need trucks, refrigeration, deals with grocery purchasers, etc. That's the actual important substance of the whole enterprise, not the system coordinating it. Couldn't you do all this on a non-blockchain computing platform, say with digital signatures, if you just got people to agree on one?
Not to mention the questionably-solvable problems of massive volatility in PolkaDots, unforseen smart contract bugs enabling and encouraging theft opportunities, constant hack attempts, potential that the whole platform could collapse if the currency collapses and nobody cares to stake it anymore.
The farm commodity hedge markets are dominated by players who leverage private access and knowledge to make massive margins. I think decentralizing this market (by anyone and their brother) would certainly benefit producers and end consumers by increasing transparency and thus lowering these transaction costs and providing liquidity. What you are talking about are auctions (like for pigs, fish or flowers) and this is already invented, centralized and corrupted and they set prices for producers and consumers in secret meetings and calls (before that wasnt even outlawed). For distribution, how easy is it to set up an ecommerce shop and for distribution you have a lot of thurd party logistics service providers. By increasing competition between these service providers of distribution etc you would again lower costs for producers and prices for consumer instead of relying on massively vertically or horizontally integrated supply chain players. Few benefits from massive monoculture productions (certainly not producers, consumers or the environment). Do you really think there are no scandals in centralized quality control systems?
Talking about price volatility. Costs for producers and prices for consumers are increasing dramatically because central banks in the western world decided to print massive amounts of currency and buying zombie companies on the stock exchange. How are these currencies considered stable in any way? Look what is happening in is happening in Turkey or Lebanon. Even worse what the US is doing to the people of afghanistan.
This is all also described in the book "Peopleware: Productive Projects and Teams". Especially the importance of individual flowtime. I tried to explain this once to an agile coach and suggested he read that book. We then continued to stuff our faces with candy while he started a "flow" game mostly focusing on imaginary situations demonstrating that collaborating as much as possible increased the flow of the team.
I think causal inference is succesfull where it is impossible do an intervention to force a randomized trail on your population. Classical statistics like in agriculture that you set up a design and field trail to find out the interactions and additive effects is sometimes not possible. Say you want to check the effect of some economic policy change or medical treatment that would be unethical to refuse to some part of the population.
I'm aware that causal inference is a popular technique in econometrics, and other places where we cannot conduct experiments.
What I'm not aware of, is if these techniques have produced highly useful and reliable inferences. (Putting on my counterfactual hat) Are there examples of observational studies that would have failed to change public policy without the techniques of causal inference?
Have a look at the literature on how smoking was established as a cause for cancer. You can't ethically intervene to have non-smokers smoke long enough to develop lung cancer. A lot of money and intellectual effort was spent on correlation not equaling causation in this case.
I'm no expert on the literature here, but Peter Norvig mentions the smoking-cancer example in his article on experiment design [0]. He gets to the same place the causality people do; observational studies.
The core idea behind a RCT is that the characteristics of a "unit" (a patient) can't affect which treatment is selected. On average, people who got treatment A are statistically the same as those who got treatment B. So you can assume any difference in outcome is a result of the treatment.
One of the simpler ways to do causal inference is by pairwise matching:
You try to identify what variables make patients different. Then find pairs of units which are "the same" but received different treatments. After the pairing process, your treatment and control groups should ("should" is doing some heavy lifting here) now be statistically "the same" by construction. Recall, that this is what we were going for in an RCT. If you did everything right, you can now apply all the normal statistical machinery that you would apply to an RCT.
The challenge is:
1. Identifying all the variables that make units alike.
2. You tend to throw away a lot of data, which reduces your statistical power. Even when the treatment classes are balanced, a given unit in class A may not pair up well with any unit from class B.
3. (Related to 2) Finding globally-optimal pairs of closest matches can be hard.
4. (Also related to 2) You need at least some people in each group. Sometimes the treatment and control are just so different that nobody pairs up very well.
In some sense, the pairing process is just a re-weighting of your data. People who are similar to someone in the other group have a large weight. People who are unlike the other group have a low weight.
You can generalize that idea a bit and reinvent what's called Inverse Propensity Score Weighting. In this case, you try to model a unit's propensity to receive a treatment, and then use 1/propensity as that unit's weight.
The intuition is: If the model says you were likely to receive treatment B (you have a low propensity for A) and you actually received treatment A, then you are likely to pair up with someone who actually received B. So we should up-weight you.
I'm currently working on my Master's thesis related to implementing propensity score matching for program evaluation in the child protection service system.
I cannot stress enough how important #1 is above. The most important part of making causal inferences in an observational experimental setting is identifying and collecting the variables associated with the treatment and outcome. It is easy to conceptualize but much harder to do in practice.
This is true only for a small subset of Causal DAGs even within this 'Causal Calculus'. It can't account for circular causality or discontinuous relationships. That's not to diminish your suggestion, only to contextualise it.
You can't really, at least not in the sense that I think most people think of it.
You basically need to make some assumptions that are broadly equivalent to assuming you've already correctly guessed certain parts of the underlying causal structure. So in a certain sense you're kind of begging the question, in a way that you wouldn't need to do if you had the ability to do interventions/randomized trials.
That being said causal inference techniques are still very valuable in making explicit exactly what assumptions you're making and how those affect your final conclusion and therefore how to minimize the impact of those assumptions.
The rules also provide a framework within which you can rule out some causal relationships. So they at least go some way to confirming which hypotheses can't be correct given the data.
Well isn't the point of ubi that people less people could work. Productivity with automation has never been higher and we still work ourselves to death for less pay. We need to take the path of increased automation anyway since we've got some sort of ubi for the elderly population already. Now it's a burden for the young people. If we could automate a lot of basic needs like food production and processing, construction, transport etc you would have unlimited near costless basic needs met and provide for the elderly. Then it would just be panem et circenses but a bit more advanced and without the slavery and the need for empire building like the romans but just technological advancement and automation.
You've got it backwards - low-skill jobs are done manually because low-skill labour is cheap. Low-skill labour is cheap because most people in it literally can't afford to not work.
We don't need to automate away all our low-skill jobs: we could just pay the workers more. Most workers are paid well below the actual value of their productivity because they have a crappy BATNA, and there's no need for employers to pay much higher than the worker's BATNA.
The more you pay the workers, the more profit in a machine that can replace min-wage workers, and the sooner a machine will be cheaper than the well-paid workers.
There is an argument that low wages hold automation back by reducing cost pressure to automate, therefore UBI, since it would increase the cost of low-skill workers, would spur automation in that sector.
I think it's a simple, logical argument, but I'm not sure if it would survive a collision with reality^^
At the risk of sounding dumb: I agree fully with UBI and automating away menial jobs that don't bring people fulfillment. But what happens once we've done that? Do the top .1% just own all of the automation equipment and thus most of the industries? Do we need some way to commoditize that aspect then, (seize the means of automation?)?
It's super hard to wrap one's head around this future where we have the challenge of preventing a few very rich from owning literally everything IMO
Well autism is sometimes labelled as individuals that are less "self-domesticated". Animals can be domesticated by humans and breeded to be tamer. Tame animals are more sociable towards humans, also have lower intelligenc and brain capacity since alot of their needs are met by the domesticator I guess. See the difference between cats/wild cats, dogs and wolves or foxes and domesticated foxes. Self domestication is the concept that we as humans have domesticated ourselves. At one end of the spectrum you have individuals that are less domesticated and fit in less with society (autistic people or psychopaths: generally lacking some sort of empathy (respectively cognitive and affective because there is a difference). On the other end of the spectrum you have Prader Willi and Angelman which are very tame and sociable but with lower cognitive abilities. https://www.psychologytoday.com/us/blog/the-imprinted-brain/...
My grandparents were farmers and used generic selection to pacify their cattle herd. It was very clear from their work that behavior was inherited. I was already predisposed to the concept. Even still it was a surprise when I found out just how much of my behavior was genetic.
money printing is done by handing it to banks at zero interest and flooded on the stock exchanges. The rich get richer and the poor can buy less with the same money since wages are the last thing to raise (while the rich buy up houses and commodities with their free money). Inflation is a tax on the poor and elderly. Literally since VAT is percentage-wise levied on relatively more expensive consumer goods and if wages rise they come in higher income pay scales for taxation. Crony capitalism at it's peak where the players get bail outs and handouts and the poor get poorer and the middle class disappears.
Mining is supposed to be increasingly cost prohibitive so in an ideal world it would eventually cease to have a meaningful impact on currency supply within practical timeframes. Assuming economic growth holds, this would mean constant deflation (i.e. value of the currency continually increases as its supply remains quasi constant while the overall economy increases).
Since transfer of cryptocurrencies incurs transaction fees and the value of those currencies only ever increases, transactions would eventually become cost prohibitive for smaller values. So realistically it would seem that we'd eventually have to transition to representative money instead (i.e. you don't transfer crypto between wallets, you shift numbers in a trusted third party's database which itself holds crypto for you). This would lose most of the benefit of the blockchain for these smaller transactions and effectively create banks because these third parties would have operating expenses they would have to charge you for (either directly or by using your crypto for investments).
So in other words, this sounds a lot like returning to the gold standard with extra steps.
that "would" in your shift argument is already reality, that's the lightning network, which transmits bitcoin off-chain for pennies, but in a decentralized way, because it utilizes existing bitcoin balances on lightning nodes for it. There is still no bank, BUT there is a more centralized "arbiter", yes. So your argument does hold about the trustlessness, though not entirely the centralization.
The one thing that's central in this debate is to understand that current currencies are built to be inflationary, which causes the need for constant economic growth. Bitcoin is not. The "bitcoin fixes it" meme is hopeful utopianism, I grant you that, but bitcoin would not carry this growth mandate in its bowels that state currencies do.
(a bit like the subprime mortgage bubble but for corporate debt)
Those faulty and badly managed companies got even more free money injected and are even more addicted to near zero interest rates because of covid...
Now if inflation becomes to high for consumers the feds have to choose between consumers or once again rescueing those companies when setting interest rates. Neither will turn out good
