After lots of investment in both time and capital I was never able to get one to work. The software stack was unsupported and after talking with everyone I could I was unable to find anyone who was able to get the software working outside of the walls of MiT.
I then worked on a related product, growcomputer (www.growcomputer.com) but ultimately there just isn't enough value in these kinds of products. As others have posted here and in previous HN posts on this topic, the A&E schools and plant scientists have long ago figured all this out, one can already purchase plant growth chambers, all manner of hydroponic and aeroponic systems, nutrients, controllers, climate controlled greenhouses, etc.
Ultimately it comes down to the economy of powering plant growth with electric lights. It just does't add up except in specific applications like mushroom farming, and very high end "foodie" niche markets. The value proposition just isn't there.
edited - formatting
There was a link 2 days ago about aerial photography of LED greenhouses in the Netherlands :
Ten acres of corn has much less value than ten acres of roses, so growing flowers is a major export industry for Netherlands.
It's difficult to compare the real-life view with pictures (just like the aurora). And it isn't always as bad. But on cloudy, cold nights, it can have an apocalyptic feel to it.
If you want to make money growing food though, mushroom farming seems like the best bet. Right now there are several edible mushroom species where, if you were to start growing them commercially, you would literally be the only person in the country offering them for sale. And some of them solve serious culinary or food science problems. E.g. right now impossible burger has been able to do a pretty good job imitating the texture of protein in ground beef, but they haven't yet been able to do a great job at imitating the texture of the fat components.
Yields are unpredictable at scale though, so it's more of a farmer's market product at the moment. It's not like morels though, where even small-scale production is very difficult.
If you're not allergic to them. I don't know that you'd really be able to sell them commercially. I don't even see them sold at farmers markets.
But I agree. I looked at this as a hobbyist (no, not for weed) and thought it could be helpful since senors and hardware are currently as cheap as they are. But industrial agriculture has complete different set of problems.
Aside from that plants in general seem to be quite resilient and don't require perfectly balanced control loops. One might think so, but your room plants constantly dying is probably just due to not watering them for months.
So in the end having a few seeds and a bit of water probably beats the food truck full of tech on all levels of efficiency.
Still an interesting topic to create these environments. But perhaps not the correct approach to solve food shortages or help with problems large scale agriculture faces.
Of course, there are the outliers (Corn is the big one I'm thinking of here) and the monocropping we tend to do is a whole different problem...
Isn't this just called a window?
If you're at the point where you're using soil from outside and sunlight from outside, to support a tiny device, why not just use a pot by the windowsill?
For many reasons: 1) Using vertical space rather than surface area; 2) Producing closer to population centers; 3) Producing year around irrespective of seasonal changes; 4) Not having to use pesticides.
1) and 2) are essentially irrelevant, because the fundamental limiting factor on both agriculture and sustainable energy generation is surface area. Plants are effectively photochemical solar arrays that turn light and CO2 into food energy; nobody would be crazy enough to propose a "vertical solar farm" where LEDs are used to illuminate photovoltaic panels, but vertical farms are exactly as bonkers.
This isn't true: at least in theory, you can take white sunlight, produce electricity by the photovoltaic process and power LEDs in a spectrum optimized for plant growth (blue and red light instead of green). In practice the LED efficiency (150-200%) doesn't yet make up for the inefficiency of the best commercially available solar panels (25-30%), but the idea is not fundamentally ridiculous.
And solar power is not always the cheapest form of electricity, or even necessarily competitive. Consider Iceland, with abundant cheap renewable geothermal energy but a shortage of sunlight, and LED-illuminated farms make even more sense (and are already operating at scale).
You are right here, however, not all plants are created equal. The value of a plant in not tied to how much energy they produce.
WRT. tomatoes, I'm not an expert, but in colder climates, don't they grow in greenhouses? Greenhouses should have zero operational carbon footprint.
 - My mind races to imagine vertical or subterranean farms powered by a nuclear plant.
 - Also known as planned obsolescence. People are assholes.
2) What's the fetish with vertical farming? Productivity of a greenhouse is 10x to 100x of a conventional field. The key limitation is capital costs. Unless you are growing small plants like basil/strawberry, vertical farming doesn't make any sense. It balloons your CapEx, increases power consumption (sunlight) and you save land, which you have saved a lot of anyway just by going to greenhouse farming from normal fields.
Do most folks not appreciate that most of the farmland we have is open field, we are not running out of space for greenhouses?
RE 2), fair. In my mind, I grouped vertical and rooftop farming together, and find them desirable in the sense of bringing back more plant life into the cities (and reducing last-mile transportation footprint). But you're right, in terms of general food production, they're not all that interesting or useful.
Is anyone aware of comparison by the numbers?
If you care about carbon footprint campaign for a carbon tax. Most of the carbon footprint of transportation is now and shall always be in the last mile, whether that’s Amazon delivering it to your door or you driving to Walmart or Whole Foods.
Capitalism is extremely good at reducing transport costs and since those costs are basically denominated in hydrocarbons it already does a better than half assed job of reducing carbon footprints. But a tax would be much better and more general.
Yes. But no transport at all is more efficient than even the most efficient method of transport. And food doesn't get straight from the docks onto the tables. There's usually lots of trucking involved. As you also say below.
> If you care about carbon footprint campaign for a carbon tax.
> Capitalism is extremely good at reducing transport costs and since those costs are basically denominated in hydrocarbons it already does a better than half assed job of reducing carbon footprints.
Capitalism is also extremely good at making the costs disappear, and reappear elsewhere, usually very diffused and paid by people not being parties to the transaction. That's why I am an advocate of carbon taxing - because it puts those costs front-and-center.
Now I'm not against shipping food, and I recognize the oceanic freight efficiency. I have two reasons for being interested in "alternative", more localized farming techniques: one, the ecological footprint (not just carbon) that's not accounted for, and two, robustness. I have this feeling that with current systems, we're one or two big accidents from a large humanitarian crisis. I feel it would be better if smaller groups of people could meet their basic sustenance needs with local produce - and ability to grow anything anywhere would definitely help with the varieties available.
Robustness and efficiency are at odds with one another. Any highly robust system will be very wasteful indeed compared to an efficient one; any system with lots of slack just isn’t efficient. Obviously if you think a large decline in standards of living is an acceptable price to pay for that robustness that’s a political position you’re free to support but good luck getting the votes for it.
Ability to grow anything anywhere requires energy. If you’re assuming civilization hasn’t collapsed anyway these farms in a box are either solving a non-problem or one with limited applicability.
RE More from Less, I need to buy and read that book. I remember an article promoting it being discussed recently on HN, and from it I got a distinct impression that US "doing more from less" is just an artifact of not accounting for embodied energy and material waste during (outsourced) manufacturing. I.e. if copper imports go down and electronic components imports go up a bit and consumer electronics production go up even more, it doesn't mean the US is better at using less copper for consumer electronics; it means you're not accounting for copper China wasted manufacturing components.
Anyway, that's my initial impression; I'll have to read the book to see if it goes into any more details to support its conclusion.
You can't make a general statement like that because it ignores the geographic advantages of the locations you are transporting from. If a certain location has a high concentration of copper for instance then its local abundance might be 1000x higher than the location you are transporting it to. Simply looking at the transportation cost doesn't capture the whole picture.
Not the best plants to grow especially in a tight space. Cucumbers will take over everything especially close together. The bush varieties are good for container gardens but the tendrils will attach to other plants. I have to keep them off of other plants or provide a trellis to grow. Last year I got an infestation of cucumber beetles. Tomatoes grow like weeds and don't latch on to neighboring plants. Both grow from seed like wild fire.
Lettuce and basil do not provide enough nutrition for this kind of setup. Enough said. Arugula is more profitable than most but will bolt with heat.
>But despite all this money and brainpower, things soon went awry in Jordan. Schroeder, in a phone interview, told Spectrum that the conditions at the NCARE site were harsh, with a very dry desert climate and high indoor temperatures. The power frequently failed, which shut down the building’s air conditioning and the food computers’ LEDs. When the air conditioning conked out, it sometimes reached 45 °C (113 °F) inside the lab.
>Worse, the Wi-Fi was unreliable. A Wi-Fi connection was necessary to remotely monitor some of the parameters inside the grow chambers, which were equipped with cameras and sensors that measured temperature, humidity, and pH levels.
I am in the Deep South and that temperature is not going to work at all. Temps in the 90s will stop my plants from growing, cause flower drop and stop fruiting but humidity here is also a factor, not a factor in Jordan.
Stable power and wifi in a refugee camp? Keep dreaming. This part should be automated with sensors and a solar battery (or backup batter in case of power loss).
I am planning a couple projects with my Raspberry Pi's to collect data and do some automation for my plants. Once I have something going I will submit a few things to HN.
Let me know if anyone wants to know more or has any questions.
Basil is also very easy to grow, so its a great beginner plant.
It should be noted that land grant schools were solely set up all over the country a hundred and fifty years ago to tackle exactly this domain of problems and not without some irony that MIT is one.
MIT's Robotic Bees will solve that problem.
It does tick all the boxes for an attractive startup though:
- Disruptive tech
- Charismatic (?) founder
- Peer-reviewed research
- Open source
- Internet of things
- Benefits the environment
- Contributes to world peace
As impractical as it is, nobody should be surprised by the fact that it got so much traction over the years. Even more far fetched projects such as roads paved with solar panels and various kinds of atmospheric water harvesters are still being actively funded and developed as we speak, despite overwhelming evidence that they are never going to be economically viable.
If you look at a Hasbro Easy-Bake Oven, it would be easy to discount "ovens" as a good way of cooking food. They're too small, and the incandescent bulbs don't produce enough heat. How could anyone thing this was a good idea?
It is...unfortunate...that the MIT name was attached to a group of people trying to pass off shoddy malfunctioning knockoffs of toys as new innovation.
The problem is this sort of matter is of course already studied by agricultural departments around the world, with a lot more competency. There is already a ton of research into, e.g. the effects of soil pH on potato production. Media Lab's food computer seems like a case of NIH, and probably wouldn't have happened if MIT had an agricultural department.
In industry you _eventually_ need to show result and earn money, it never get to that part in academia.
People publish papers that are practically useless but can be cited by other people in the same position, which then improves their standings
There's a sucker born every minute and a skilled con man can exploit that unfortunate truism to keep a scam for quite a while. Moller's skycar is an example of a flagrant scam that lasted for decades.
> It doesn't take long to think of entire industries that are shady, from payday lenders
I trust you aren't arguing that payday lenders don't earn money!
There's a huge difference between "I don't like what this industry is doing" and "this industry is fraudulent, can never produce the goods/services claimed, and will never turn a profit for the investors".
The immorality of Pay Day is very much known, and it is not relevant because Pay Day is scamming their customers, not their investors.
The rideshare industry famously doesn’t make a profit.
Neither does the electric scooter industry.
The meal in a box industry has retention problems, because eventually their customers learn to shop and cook.
Electric scooters are a little iffier, as they take significantly more capex and there isn't much data on market acceptance (Or, at least, none of which I know.)
Meal-in-a-box is not a bad idea either; new people are showing up every day whining about how, "Adult-ing is hard." The fact that people eventually "graduate" is not an issue so long as more people are being born. It's just not necessarily something Joe Six-Pack is going to buy.
If you've noticed a common theme here, it's that the idea of hyper-scaling can be flawed. The above all have significantly lower threshold for market saturation than they believe, if you ask me.
Ride-sharing can be successful, but may not take over the world. Electric scooters can be successful, but may not take over the world. Meal-in-a-box can be successful, but may not take over the world.
SV gets stuck in the pareto trap of scaling past economies of scale and to the point where that next 20% of your customer base takes 80% of your capital. You can push scaling, but the market only moves so fast.
Ditto for electric scooters - I rarely see anybody use one but it is somehow already costing my country millions in medical costs.
Like Whole Foods or broadcast.com?
Edit: Not such a good idea in 1999, when operating without economies of scale and scope, and with high transportation costs.
But this is stupid levels of over-hype, and it has nothing to do with academia / business. It has everything to do with overbranding and rushing the hype.
By the way, the "research" about texting while driving you might have read about or heard discussed on NPR is just highly plausible bullshit.
Sure, Springer is making money, but I never met — nor heard of — anyone in academia making money through publishing. The truth may come as a surprise to you, but publishing articles, reviewing articles, editing journals, all of that is done for free.
You might make some money through consulting, but publishing? No. Not even when you write a textbook.
But _sometimes_ it’s entirely not warranted, the party just wanted the name attached. Just like how company brings in a consulting firm to confirm its existing decision.
Which sometimes buffles me because it doesn't reward particularly well for a position reserved to people with strong intellectual prowess, it merely prevent people from being fired for doing nothing.
There are definitely people in academia that are driven for truth, but I don't know if that's the majority or a minority.
But academic projects are inherently closer to aspirational, so I'm not sure it's fair to conflate aspirational with fraud. This episode sounds like fraud. Aspirational is claiming your novel magnetic media "could" enable a terabyte per centimeter storage density when you know perfectly well it couldn't hit that in a real application. I dislike the rhetorical overreach and would not use naive statements like that, but it is a different thing entirely to take claim for work that wasn't even done.
This is true for every major industry and the political class and has been true for decades.
Eric Drexler, the "Engines of Creation: The Coming Era of Nanotechnology" author, got his PhD at the MIT Media Lab.
>His Ph.D. work was the first doctoral degree on the topic of molecular nanotechnology
>futurist thinktanks/institues and similar
Sounds like a good fit for media lab since it's all fluff.
Let's not throw the baby out with the bathwater. All of the things on this list are desirable and somebody has to do research for us to ever achieve them. The problem still is people trying to misrepresent the state of the art, whether in academia or in the industry.
I may be wrong.
Feeding the hype machine that generates donations is considered more important than the work (not to mention the truth-seeking function of academia) the donations are supposed to fund.
There needs to be some serious house-cleaning if they want to be considered an actual research lab of any repute. Ito was a problem, but he quite clearly was not the only one.
What academia should not be divorced from is knowledge and truth seeking. I don't mean that in some wooly, idealistic sense: if that isn't the focus, you're hucksters and frauds, not academics.
[ETA: a large number of media lab projects are still really good, as the other commenter mentioned]
At least one other major project, the OLPC from media lab and the infamous Negroponte gang was a total failure. Negroponte was back in the news recently due to the whole Joi Ito-Jeffrey Epstein pedo funding debacle.
Or most of the rest of the stuff from that group: http://cameraculture.media.mit.edu/research/
Source: I was there for the worst two years of my life.
"Why MIT Media Lab thought it was doing right by secretly accepting Jeffrey Epstein’s money"
The talks I've seen out of TEDx tend to be more "this is interesting" rather than "this is 100% the truth about ____". I think most TEDx events take themselves less seriously, which results in fewer speakers that think they are the end-all-be-all of some topic. Where as TED events seem to try to get the "foremost expert on ____", even if the entire area of _____ is mostly quackery.
Though really I think the pompousness is proportional to size and has (almost) nothing to do with TED vs TEDx. Anecdotally, the fairly large TEDxGlasgow had a lot more wankiness than the quite small TEDxUbud.
BUT, come on folks, refugees are not your fucking PR opportunity. I can't imagine how insulting it must be to need food, shelter, and a source of income, and some clueless techbro from across the globe sends you a malfunctioning Raspberry Pi grow tent. How disconnected from reality does one have to be to think this is a good idea.
But this "food computer" is what happens when you think smart people are smart with everything or that they know everything. They aren't and they don't.
"MIT Built a Theranos for plants" is a headline that's true in a lot of levels (further down in the thread)
(Not to mention all the stuff about reputation laundering)
I really can't stand the elite university system. It's not that these schools are bad, just that many others are also good but don't have the brand.
Now, most of it is proprietary, expensive, and the control systems are fairly basic, so there is scope for an open source project, for improvement, for openness and usability.
Or are you just saying that low-end growth chambers also exist that don't include "food computer" features?
Unlike Engineering, there is no Board of Engineers to hold you (or your company) to account for lying about the capabilities of your core product or whether its actually working for the vulnerable people used in a company's marketing :c
We know categorically that hand cranks do not work, although I think Thad Starner (also from Media Lab) had already proven that when the project started.
They were trying to make them homogenous, when probably a pair or trio of devices would have worked better and had higher survivability.
I hope someone more sober tries again.
This is exactly the ethos of move fast and break things that will make the right things happen. They shipped to real customers in real environments. They used technology to address problems. They iterated on the design. All just like you're supposed to do for any successful product.
Where they screwed up is claiming to "make the world a better place" before they really did it. They're still in product-market fit phase, not techbros save the world phase.
Or, they wasted fuel, energy, manpower and hopes.
> They used technology to address problems.
Except they didn't, beyond noticing the problem exists.
> They iterated on the design.
Everyone does that.
> All just like you're supposed to do for any successful product.
The first rule for a successful product is, or at least should be, that it does something useful. The second, that it does it economically in context of available alternatives.
"Make the world a better place" is cringe-inducing, but reducing the message slightly is still bullshit marketing. For me, this already ruined any trust I had in the past in MIT ML, TED talks, and countless of startups "saving the world", "improving lives", etc.
(Also, a thing I learned from observation: if you want to improve the world, you probably don't want to take VC money.)
It's absolutely disgusting to make a living for yourself by pretending to help refugees and lying about it.
Using refugees to push your nonfunctinal product is slimey to say the least.
Sounded like a cool idea... too bad Caleb is a scammer.
By "agrifood commodities" do you literally mean "vegetables or cereals"?
People all over the world have been practicing agriculture for thousands of years. Generally it's more effective using things like spades, sickles, etc. than a 3D printer-looking nonsense.
There is also a misconception about "shade plants". All of the major plant crops that I know of want full sun. Sunburn in avacodos (for example) is due to pruning or exposed stem/bark and the reason why paint (er, sunblock) is applied is an efficiency measure; the exposed stem also stimulates fruit formation so you don't want too many leaves (which naturally would prevent sunburn).
Nethouses (vs. Greenhouses vs. Glasshouses) are common in areas with a good climate (such as Almeria in Spain) but they are not always the optimal solution. The further requirement to justify nethouses is space as a limiting factor (or water).
But I would categorise Nethouses still as being "outside farming". They don't even stop rain from getting through.
> the size of a dorm-room refrigerator
It's simply a joke or a hobby for rich westerners in a city apartment.
But the real question is: to what extent is this a systemic issue? In other words, does the current system/ environment select for these types of these PR over engineering people?
Next time these cons will sell 'save the planet with geo-engineering' to another gullable audience and we'll all be screwed big-time.
Another title could have been: Nonprofit sends food growing technology to refugee camp. That's a pretty generous use of the word "used".
Should they have intentionally kept silent about the fact that they were actively working with refugee camps until they had peer reviewed proof that it worked?
No one here did anything wrong. The technology failed. They will either iterate or decide the idea isn't worth pursuing. No outrage necessary.
I’m very happy to read that some science teachers in the US get it right.
Imagine that applied to today.
Well, today, it would be a start up claiming to have secured telnet, with a webapp, carry over session between mobile and desktop, and gathering telemetry between each connection.
(apologies, this has turned into my yearly rant of "make protocols, not apps!")
I wouldn't be surprised if there are more, just with a different phrase.