Hacker News Reader

Comment by mberlove | Hacker News Reader

"Farms are now complex technical operations that use networks of remote monitors that measure soil moisture, or robotic milkers that can detect an infection in a single teat, or environmental control systems that maintain the precise indoor temperature and air filtration needs of a poultry barn."

Both a completely understandable and even desirable outcome of advancements in animal husbandry, and an unacceptably vulnerable solution to a massively important system where low-tech solutions are well available.

Modern massive farming requires tech to be efficient? That feels like a reasonable statement though I have no proof of it. Maintaining or improving farm output at scale requires massive centralized farming? That feels like less of a reasonable statement, and again, I have no proof one way or another.

Are there studies on this? Can we decentralize, downscale, without compromising outputs?

Just ruminating here (pun definitely intended).

View on HN

Pretty much every production improvement in either farming or manufacture since the 19th century has been:

- scaling up is more efficient, because the production increase has a better "slope" than the increased capital investment

- increasing uniformity is more efficient, through the means of statistical process control, which requires more information and information processing

Spoke to a few farmer friends. They tell me that large scale operations are very technically complex and have many digital systems deployed for remote management. That being said, a lot of the numbers here seemed to be miss reported. This seems like a combination of genuine issue and fear pandering.

Is there any goods-producing industry that is more efficient by being decentralized? This isn't a rhetorical question, I'm really not sure.

And I don't mean software, media, IP, etc. but stuff like milk or meat or lumber or micro chips.

Ginseng is an interesting one [0]. It can be farmed without too much trouble, but farmed ginseng has a more uniform, smoother appearance than wild ginseng. One of the major uses is in Chinese traditional medicine, where it's believed that wild ginseng has better properties (exactly what properties and how you measure them is hard to pin down).

This leads to a tension: You can set up an efficient, high yield, centralized farm, but your main buyers will value your crop less. Or you can use "simulated wild" conditions that create a wild-like appearance and charge more, but now you need a very different operation that looks more like forestry than farming and is much more likely to be decentralized. And of course you can still forage, but that has its own limitations (lots of details in [0]).

[0]: https://www.uky.edu/ccd/sites/www.uky.edu.ccd/files/ginseng....

Certain rapid prototyping can be faster by being decentralised, but this results in higher per-unit costs. It's cheaper to 3D print than do injection moulding if you just want one item, but not if you want 10,000.

You know the old saying in computing, "Fast, Cheap, Good - pick two"?

That applies to farming too.

If you're okay with paying roughly 50% more across the board in the supermarket, then farming can be rather less dependent on technology.

Also if you are okay with many starving to death. There are more then 8 billion people on earth who need to eat

If you're going to do the high tech intensive route, though, then you must needs also have a security expert on staff, and I bet you that's a rare thing.

Forget a security expert, from what I've witnessed in a lot of industries security can be improved 100 fold by simply listing to what any half decent IT staff would recommend. So many business still think its ok to have a single login off a vital system with a 4 character password shared amongst 30 employees.

Reply on news.ycombinator.com