Hacker News new | past | comments | ask | show | jobs | submit login

True historical research isn't something many undergraduates do, and I can see how it would be exceptionally valuable.

Unfortunately, many undergrad courses simply have you read existing analysis, or at best present you with "primary sources" that are served up to you with no real digging around. It's not that the books you read are of low quality - often, they're really high quality. I remember reading a very elaborate history about the naval arms race between Britain and Germany prior to WWI, and I got a lot out of it. But what I never did was truly dig into the data. I believe it, but did I ever actually go to the the documents (without someone spoon feeding them to me in a way that leads to only one conclusion)? Nope.

Here's the thing - this can actually lead to big errors later in life. One amusing story I had - I worked on supply chain issues for a big manufacturing company once, and during an interview, I was talking excitedly about various mathematical approaches. The interviewer (a math PhD who eventually did hire me) stopped me and reminded me of how detailed the work really is. He said that one young engineer who worked there had spent six months digging through documents, and had discovered the source of an inefficiency. The QA tests for some components was different in Asia and Europe, so the pieces were passing in one location, getting shipped, failing in the other, getting shipped, passing… in this loop. It had nothing to do with math, it had everything to do with actually getting truly ground level with documents, tests, bits of paper, factory floors, and figuring out what exactly was actually going on in the world.

On a grimmer note, a huge factor in the banking crisis was that people used mathematical models without actually looking at the reality behind them. Yes, a package of very high quality mortgages has a low probability of collapsing, and if you take that low probability number and stick it into your model, all is well. But if you actually dig into the reality, find the documents, realize many of the documents don't exist, and realize that the ones that do are not at all the high quality loans that would typically get a strong rating, and ask why, realize that the package was rated with stronger loans that were taken out replaced with bad loans, the new ones being used to create a new package with a new high rating, over and over, and that the company that did this is now taking out insurance policies against the loan packages that they prepared because they know they have a vastly higher rate of default than the insurers think (based on the faulty rating)…

well, then you're thinking like a history major who actually dealt with true historical research. And the few people who did this were aware of what was coming long before everyone else.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: