home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.physics
- Path: sparky!uunet!snorkelwacker.mit.edu!galois!riesz!jbaez
- From: jbaez@riesz.mit.edu (John C. Baez)
- Subject: Re: Symmetries, groups, and categories
- Message-ID: <1992Aug21.181701.27612@galois.mit.edu>
- Sender: news@galois.mit.edu
- Nntp-Posting-Host: riesz
- Organization: MIT Department of Mathematics, Cambridge, MA
- References: <1992Aug17.184957.18492@galois.mit.edu> <1188@kepler1.rentec.com> <mcirvin.714344140@husc8>
- Distribution: na
- Date: Fri, 21 Aug 92 18:17:01 GMT
- Lines: 38
-
- In article <mcirvin.714344140@husc8> mcirvin@husc8.harvard.edu (Mcirvin) writes:
- >andrew@rentec.com (Andrew Mullhaupt) writes:
- >
- >>Hey while we're hammering physics for ill-chosen terminology, can we at least
- >>change it to 'renormalization monoid'?
- >
- >And "pseudometric tensor" in relativity.
- >
- >My algebra's rusty. What's the difference between a monoid and a
- >semigroup? Some textbooks say that it's really a renormalization
- >semigroup. They give the nature of the composition law as evidence.
- >"Monoid" is more general, right?
-
- I believe that there are some difference in opinion about these
- definitions in the math literature, but the most widely used definitions
- these days are as follows. A semigroup is a set with an associative
- binary operation. If it has an element that is both a right and left
- identity, it is called a monoid. (Easy to remember, because the "mono"
- refers to the identity, or "unit".)
-
- There used to be a rock band at Princeton called Monoid, because it was
- a "semigroup with an identity".
-
- >What I find strange is that the term is used even though the
- >[monoid/group/semigroup] nature of the beast isn't particularly
- >illuminating or important. The term seems to be the result of
- >enthusiasm for group theory bleeding into places where it has
- >little relevance.
-
- Well, scaling is a monoid (the real numbers in (0,1] form a
- monoid under multiplication), or if you allow "scaling up," a group (the
- real numbers > 0 form a group under multiplication). But I am probably
- one of millions of mathematicians who were sorely disappointed to find
- out that the "renormalization group" was such a boring group. What's
- interesting is not this group per se but its ACTION on the physical
- system one is studying (the "renormalization group flow" on some space
- of coupling constants). So I agree that it's a bad term.
-
-