JoshuaZ comments on How minimal is our intelligence? - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (214)
Well, no. In modern times number theory has been extremely relevant for cryptography for example, and pretty much all e-commerce relies on it. But other areas of math have direct, useful applications and have turned out to be quite important. For example, engineering in the late Middle Ages and Renaissance benefited a lot from things like trig and logarithms. Improved math has lead to much better understanding of economies and financial systems as well. These are but a few limited examples.
You are missing the point in this context having the taboo against book burning is helpful because it is something one can use as a warning sign.
So I'm curious as to how you are defining "good" in any useful sense that you can reach this conclusion. Moreover, the sort of thing that Roth does is in the process of being more and more useful. His work allowing for organ donations for example not only saves lives now but will go on saving lives at least until we have cheap cloned organs.
This is wrong. His work with malaria saves lives. His work with selling computer software involved making mediocre products and making up for that by massive marketing along with anti-trust abuses. There's an argument to be made that economic productivity can be used as a very rough measure of utility, but that breaks down in a market where advertising, marketing, and network effects of specific product designs matter more than quality of product.
Yes, to the point where I have to wonder how drastically far off our unstated premises about the world are. If anything, it seems like we have the exact opposite problem. We have a massive oversupply of "quants" and the like who aren't actually producing more utility or even actually working with real market inefficiencies but are instead doing things like moving servers a few feet closer to the exchange so they can shave a fraction of a second off of their transaction times. There may be an "oversupply" of how many academics there are compared to the number of paying positions but that's simply connected to the fact that most research has results that function as externalities(technically public goods) and thus the lack of academic jobs is a market failure.
No-one is disputing that mathematics can be useful. The question is, if we had slightly more advanced number theory slightly earlier in time, would that have been particularly useful? Answer - no.
No, I am not missing the point. I am perfectly willing to concede that a taboo against book-burning might be helpful for that reason. But here we have an example where people were,at the same time as burning books, doing the exact worse stuff that book burning is allegedly a warning sign of. But no-one complains about the worse stuff, only the book burning. Which makes me disbelieve that people care about the taboo for that reason.
People say that keeping your lawn tidy keep the area looking well-maintained and so prevents crime. Let's say one guy in the area has a very messy lawn, and also goes around committing burglaries. Now suppose the Neighbourhood Watch shows no interest at all in the burglaries, but is shocked and appalled by the state of his lawn. We would have to conclude that these people don't care about crime, what they care about is lawns, and this story about lawns having an effect on crime is just a story they tell people because they can't justify their weird preference to others on its own terms.
Or, we could just allow a market for organ donations. Boom, done. Where's my Nobel?
Now, if you specify that we have to find the best fix while ignoring the obvious free-market solutions I don't deny that Alvin Roth has done good work. And I'm certainly not blaming Roth personally for the fact that academia exists as an adjunct to the state - although academics generally do bear the lions share of responsibility for that. But I am definitely questioning the value of this enterprise, compared to bringing cheap food, clothes, etc, to hundreds of millions of people like Sam Walton did.
I don't see why "saves lives" is the metric, but I bet that Microsoft products have been involved in saving far more lives. Moreover, people are willing to pay for Microsoft products, despite your baseless claims of their inferiority. Gates's charities specifically go around doing things that people say they want but don't bother to do with their own money. I don't know much about the malaria program, but I do know the educational stuff has mostly been disastrous, and whole planks have been abandoned.
Obviously very far indeed.
Answer: Yes. Even today, number theory research highly relevant to efficient crypto is ongoing. A few years of difference in when that shows up would have large economic consequences. For example, as we speak, research in ongoing into practical fully homomorphic encryption which if it is implemented will allow cloud computing and deep processing of sensitive information, as well as secure storage and retrieval of sensitive information (such as medical records) from clouds. This is but one example.
Well, there is always the danger of lost-purpose. But it may help to keep in mind that the book-burnings and genocides in question both occurred a long-time ago. It is easier for something to be at the forefront of one's mind when one can see more directly how it would have impacted one personally.
So, I'm generally inclined to allow for organ donation markets (although there are I think legitimate concerns about them). But since that's not going to happen any time soon, I fail to see its relevance. A lot of problems in the world need to be solved given the political constraints that exist. Roth's solution works in that context. The fact that a politically untenable better solution exists doesn't make his work less beneficial.
So, Derstopa already gave some reasons to doubt this. But it is also worth noting that Walton died in 1992, before much of Walmart's expansion. Also, there's a decent argument that Walmart's success was due not to superior organization but rather a large first-mover advantage (one of the classic ways markets can fail): Walmart takes advantage of its size in ways that small competitors cannot do. This means that smaller chains cannot grow to compete with Walmart in any fashion, so even if a smaller competitor is running something more efficiently, it won't matter much. (Please take care to note that this is not at all the mom-and-pop-store argument which I suspect you and I would both find extremely unconvincing.)
Ok. Do you prefer Quality-adjusted life years ? Bill is doing pretty well by that metric.
"Involved with" is an extremely weak standard. The thing is that even if Microsoft had never existed, similar products (such as software or hardware from IBM, Apple, Linux, Tandy) would have been in those positions.
Let's examine why people are willing to do so. It isn't efficiency. For example, by standard benchmarks, Microsoft browsers have been some of the least efficient (although more recent versions of IE have performed very well by some metrics such as memory use ). Microsoft has had a massive marketing campaign to make people aware of their brand (classically marketing in a low information market is a recipe for market failure). And Microsoft has engaged in bundling of essentially unrelated products. Microsoft has also lobbied governments for contracts to the point where many government bids are phrased in ways that make non-Microsoft options essentially impossible. Most importantly: Microsoft gains a network effect: This occurs when the more common a product is, the more valuable it is compared to other similar products. In this context, once a single OS and set of associated products is common, people don't want other other products since they will run into both learning-curve with the "new" product and compatibility issues when trying to get the new product to work with the old.
That some people make noise about wanting to help charity but don't doesn't make the people who actually do it as contributing less utility. Or is there some other point here I'm missing?
Yes, there's no question that the education work by the Gates foundation has been profoundly unsuccessful. But the general consensus concerning malaria is that they've done a lot of good work. This may be something you may want to look into.
My answer is "probably yes". Mathematics directly enables entire areas of science and engineering. Cathedrals and bridges are much easier to build if you know trigonometry. Electricity is a lot easier to harness if you know trigonometry and calculus, and easier still if you are aware of complex numbers. Optics -- and therefore cameras and telescopes, among many other things -- is a lot easier with linear algebra, and so are many other engineering applications. And, of course, modern electronics are practically impossible without some pretty advanced math and science, which in turn requires all these other things.
If we assume that technology is generally beneficial, then it's best to develop the disciplines which enable it -- i.e., science and mathematics -- as early as possible.
He was talking about number theory specifically, not mathematics in general -- in the first sentence you quoted he admitted it can be useful. (I doubt advanced number theory would have been that practically useful before the mid-20th century.)