4 ms·
It seems to me that the definition presented there is relatively free of jargon, except the "in other words" section which links it to other (linked) structures
by cmrx64 11y ago
It seems to me that the definition presented there is relatively free of jargon, except the "in other words" section which links it to other (linked) structures.
- gh02t 11y ago> except the "in other words" section which links it to other (linked) structures. That's what I was getting at. I took a few classes in AA and its sibling topology, including at graduate level. Most of the time was spent defining things, because there's so many terms with very specific definitions. Only after maybe 2/3rds of a course do you really start getting into deeper material. In my brain, reading definitions was always a process like: read through, substitute definitions for jargon terms, read through, substitute... iterate until you hit a fixed point. That's standard for math, but AA is particularly well known for its colorful choice of names. Some people enjoy it, but I was always more attracted to the analysis-family areas of math. I've always found the other name for category theory amusing though: https://en.wikipedia.org/wiki/Abstract_nonsense https://en.wikipedia.org/wiki/Abstract_nonsense
- cmrx64 11y agoFair enough! My biggest struggle taking AA was also remembering the definitions of everything. There are so many things all tied together.