Don't Read History for Lessons | Hacker News Reader

Don't Read History for Lessons

commoncog.com

54 pointsivank10 comments

> In a sentence: I think we should read history for concept instantiations, not lessons.

By this, the author seems to mean reading for patterns instead of rote facts. And if you go to anyone teaching history in any college anywhere, they'll tell you something similar.

If only this autodidact had studied history . . .

The article isn't about reading for facts vs. patterns. Rather, the author's point is that you should be very careful about identifying these patterns, otherwise you'll likely overfit.

> In response, some folk back into the weak form of the argument. The weak form is what you get when you hear someone say “history doesn’t repeat itself, but it rhymes” — that is, perhaps the lessons of history are too specific to a particular context to generalise, but that some … similarity is consistent over time

Well, it seems you're drawing on the same argument that has already been addressed (or at least mentioned in passing) in the article.

So ... pop quiz: how is reading for concept instantiations different from reading for patterns?

Lousy article. For example, the author complains about a supposed 'lesson learned' that says (shortened) "If so-and-so happens it may be because of A, or it may be because of B". His complaint is "except that it may be because of C or D".

But the 'lesson' he presented never claimed to be an exclusive either-or. It just says "here are two possible explanations for something".

There's sloppy thinking like this throughout. Thumbs down for me.

When I read this type of stories I cannot help but think that the world would be better off if we banned that type of predator-like behavior, those super aggressive strategies that aim at getting the concurrence out of business by lowering the price more and more until you get a monopoly.

No-one can deny that what he did was a huge innovation, the fact that it is still being used 60 years later is a testament of that, but how much innovation was killed in the egg when competitors were forced out of the market?

Well, you could look at it the other way. How much innovation was enabled when everyone in the world who wanted to build an electronic gadget had the parts available earlier for less money?

You see less supplier innovation. I see more customer innovation.

If the innovation is worth something, even if the company is forced out of the market, it can be either reused in further endeavours of the shareholders/company owners or the research can be continued under the guidance of some other company. Businesses don't just "forget" about their past research (unless we're talking trillion dollar companies).

Good article, bad title.

Agree. When I went to read it, I expected a general discussion about the fallacy of using history to predict the future. (History is an excellent way to learn about human behavior. That alone is a great reason to study it.)

You can't really learn much from the history of a specific (and remarkably talented) individual.

When I go back to the top list like "1. Good managers cannot ‘manage anything’. Specialisation counts", it seems like this is the sort of conclusion you can also draw if you have multiple "prototypes" in your head, which come from reading more history. (Whether they're actually true or not, it seems.) So, the conclusion here is "don't extrapolate from a dataset of 1". It seems like this was an awfully roundabout way of getting there.

Reply on news.ycombinator.com