> Sure, but in the old system people just starved to death when there were problems with their crops (Irish potato famine, dust bowl, etc.).

Sure, but then we're trading smaller, more frequent disruptions at the cost of risking less frequent, but much larger disruptions. More to my original point though (I may have rambled there and not been clear), the risk I'm raising is that we now make decisions based on only today's situation and are unaware of the context that got us here. That is fine most of the time, but incremental change isn't fool proof and sometimes the context of how you got here is extremely important in making the next decision.

> people who care can find out more than any farmer 100 years ago about food and source theirs accordingly.

There are a few risks there though, maybe they're worth it but still risks.

You don't know what you don't know, and in that case its hard or impossible to find it online. Plenty of historical knowledge also doesn't live online at all, its still hard to find research papers more than a few decades old - at best they're online as a PDF and likely not indexed or searchable.

We also can't expect those in charge to know much of anything when the scale of lost, unknown context grows too far and too fast. At best they outsource that knowledge to others, but those are likely experts only in one small piece of the puzzle. To me that seems like a very delicate balance that can work for a time bit would inevitably fail in ways we couldn't predict.

All that said, I'm also not trying to make the argument that we must know all the context and history of anything we deal with. Just the importance of at least recognizing what we don't know and where the risks are.