home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!stanford.edu!CSD-NewsHost.Stanford.EDU!Sunburn.Stanford.EDU!pratt
- From: pratt@Sunburn.Stanford.EDU (Vaughan R. Pratt)
- Subject: Re: Report on Philosophies of Physicists
- Message-ID: <1992Sep15.050118.15796@CSD-NewsHost.Stanford.EDU>
- Sender: news@CSD-NewsHost.Stanford.EDU
- Organization: Computer Science Department, Stanford University.
- References: <1992Sep11.152100.438@guinness.idbsu.edu> <TORKEL.92Sep13094850@bast.sics.se> <1992Sep15.035832.7576@cbnewsm.cb.att.com>
- Date: Tue, 15 Sep 1992 05:01:18 GMT
- Lines: 48
-
- In article <1992Sep15.035832.7576@cbnewsm.cb.att.com> mls@cbnewsm.cb.att.com (mike.siemon) writes:
- > - it is VERY important to "break" students from reliance on the
- > geometry and algebra of their earlier studies as *the* only
- > matter of a generalized "geometrical" reasoning. There is such
- > a wealth of counter-example to expectations in the basic stuff
- > of a point-set topology course as to open previously closed minds.
-
- This familiar assertion about topology should be contrasted with the
- following conclusion to the prefatory "Note to the Reader" from
- Munkres' "Topology", Prentice-Hall 1975.
-
- Now it is all too easy in studying topology to spend too much
- time dealing with "weird counterexamples." Constructing them
- requires ingenuity, and is often great fun. But they are not
- really what topology is all about. Fortunately, one does not
- need too many such counterexamples for a first course; there is
- a fairly short list which will suffice for most purposes. Let
- me give it here:
-
- [product, uniform, box topologies on R^n
- R with basis all [a,b)
- the order type S_Omega (better known to logicians as omega_1)
- the lexicographic square]
-
- I would add to this perspective that the "weird counterexamples" (at
- least the ones I vaguely recall being put through) are for the most
- part the *fault* of a set-theoretic perspective. Topological spaces
- and their continuous maps don't form a closed category (one such that
- the set of continuous functions from S to T forms a topological space
- in the "obvious" way), forcing one to jump around between spaces and
- sets when thinking about maps f:SxT->U and fixing a point in S or T.
- For ordinary consumers of topological spaces it's healthiest to go
- right to compact Hausdorff spaces without messing around with
- non-Hausdorff or non-compact counterexamples---not only is CGHaus
- closed but very remarkably it is algebraic, i.e. essentially a variety
- or equational class, just like groups or lattices, only with operations
- at arities of every cardinality. For those needing the whole of Top
- but cleaner, locales are the clean way to go. (Cleanliness is next to
- godliness. God I hate it when I get religious.)
-
- Mathematicians who don't see the connection of their stuff with
- foundations are muscle-bound programmers who write large grungy codes
- without even being aware there's a problem.
-
- --
- ======================================================| God found the positive
- Vaughan Pratt pratt@cs.Stanford.EDU 415-494-2545 | integers, zero was
- ======================================================| there when He arrived.
-