My favorite thing about the Monthly are the problems. I particularly like it when it goes like this: Problem X: [Proposed by someone] [Problem statement] Solution: [Solution by Proposer] Editors note: Several solvers noted that this problem already appeared in this Monthly, as Problem Y... Then you go to look at the earlier Problem Y on jStor and find Problem Y: [Proposed by Someone Else] [Problem statement] Solution: [Solution by Someone Else] Editors note: Several solvers noted this problem already appeared in this Monthly, as Problem Z... Don Knuth observed that chess community has a pretty good institutional memory for games---you always see things like Alekhine-Tartakower (place) (tournament)--- and it's now pretty easy to find the games in databases. But for math problems everyone seems to be solving the same ones over and over again and there's no standard system of referring to them. Problems get scattered into textbooks, exams, periodicals, math olympiad texts, etc with a lot of duplication. The OEIS shows the value of getting systematic about this