Chaff? For a project where one has total control, yes. When one has to get approval and deal with a specific list of approved licenses (which may not include one's particular favourite, no matter how sensible it seems to oneself), then no. It is appreciated.
Bandits are great, but using the theory correctly can be difficult (and if accidentally incorrectly applied then ones results can easily become pathologically bad). For instance, the standard stochastic setup requires that learning instances are presented in an iid manner. This may not be true for website visitors, for example different behaviour at different times of day (browsing or executing) or timezone driven differing cultural responses.
There is never a simple, magic solution for these things.
The truth of the essay is illustrated by studies that show lower social mobility over longer generational windows, i.e. ones grandfather's income quintile is a better predictor of ones owns than ones fathers's. The intangible cultural inheritance of social class is more lasting than fluctuations in material wealth.
I've always found the reflexivity of class to be somewhat interesting. Economically, my family is in the bottom quartile of the country—but it never felt like that. They definitely raised me with "upper class" values and attitudes. (For example, there was never even a shadow of a doubt about whether I'd attend college.) My grandparents, on the other hand, were quite well off.
As such, I've always felt more comfortable and fit in with people whose families have incomes vastly exceeding my family's.
> "ones grandfather's income quintile is a better predictor of ones owns than ones fathers's"
Wouldn't this actually meant that the average of one's parents, grandparent's, and great-grandparent's income quintiles is a better predictor than one's parents alone? (Unless these studies only looked at personal income of male relatives instead of household income?) The income of the grandfather is still a single point subject to fluctuations.
What a strange statement/article. I had to check all three of those as an undergrad, associativity of matrix multiplication was definitely first year linear algebra, and I think elliptic curve product was first year number theory.
No, Chinese characters do not stop "moveable type taking off". Moveable type printing was in fact invented in China, nearly 1000 years ago, and widely used.
Movable type actually proves the author's point. It was an amazing idea which didn't take off in China as fast as it did in the West, precisely because it was a huge hurdle to manufacture the ~1000s of blocks required to represent each Mandarin word.
See Wikipedia [1]. Movable type was first invented in China around A.D 1040, about four hundred years earlier than in the western world.
In Qing Dynasty, the government used it to print 64 sets of the encyclopedic Gujin Tushu Jicheng. Each set consisted of 5040 volumes, making a total of 322,560 volumes printed using movable type.
It should be noted that this Qing Dynasty encyclopedy was printed in 1726, seven centuries after the invention of movable type and more than 270 years after Gutenberg's Bible. It is also a Government sponsored project and probably wasn't cheap. The point still stands that compositing a book was far more labor intensive with chinese characters than with any alphabet. You need a few dozen types for an alphabet and at least 100 times more fore Chinese. An alphabet makes it considerably cheaper and not reserved to a tiny elite with 64 copies. That's the main point. Common people couldn't afford it.
There is no doubt possible that using an alphabet greatly helped the dissemination of knowledge. Between 1455 and 1500 there were over 30,000 distinct incunables edited in europe. Not 30,000 copies. More than 30,000 different books. [1]
Gutenberg's press went viral [2] and vastly more widespread than what was seen in Asia.
In the 15th century Korea adopted Hangul, the Korean alphabet.
"A potential solution to the linguistic and cultural bottleneck that held back movable type in Korea for 200 years appeared in the early 15th century—a generation before Gutenberg would begin working on his own movable-type invention in Europe—when Sejong the Great devised a simplified alphabet of 24 characters (hangul) for use by the common people, which could have made the typecasting and compositing process more feasible. Adoption of the new alphabet was stifled by Korea's cultural elite, who were "appalled at the idea of losing hanja, the badge of their elitism."
"It was effective enough at disseminating information among the uneducated that Yeonsangun, the paranoid tenth king, forbade the study or use of Hangul and banned Hangul documents in 1504,and King Jungjong abolished the Ministry of Eonmun (governmental institution related to Hangul research) in 1506.
The late 16th century, however, saw a revival of Hangul" [3]
> Moveable type printing was in fact invented in China
I know that's my point exactly.
It was definitely not as widespread and explosive as Gutenbergs printing press though. The number of characters is two orders of magnitude smaller and cuts down the manual labor far more effectively. You can't argue against that.
Woodblock printing was in use long after the invention of movable type because of that.
Very neat. One problem is immediate: when I handwrite an "x" for maths, it is as two 'c's touching back-to-back, so as not to confuse with some multiplication/cross-product symbol; such input proves very confusing to this code.
> so as not to confuse with some multiplication/cross-product symbol
There really isn't much need for the multiplication symbol when we have side-by-side multiplication notation, and if x is a vector then you can underline it to distinguish it from the cross product (or just not name your vectors x). The "two c's" x has always looked bad to me and ends up getting confused with α and other things.
I usually use an italicized x for math variables, a dot for multiplication when not implicit in the formatting of the expression, and an uppercase X for cross product.
