Article 71295 of sci.math: Path: news.sri.ucl.ac.be!jussieu.fr!math.ohio-state.edu!uwm.edu!vixen.cso.uiuc.edu!newsfeed.internetmci.com!info.ucla.edu!nnrp.info.ucla.edu!ewald.mbi.ucla.edu!laura From: Laura Helen Newsgroups: sci.math Subject: Re: Sci.math.grad -- preliminary proposal Date: Sun, 10 Mar 1996 13:54:36 -0800 Organization: University of California, Los Angeles Lines: 20 Message-ID: References: NNTP-Posting-Host: ewald.mbi.ucla.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: On 10 Mar 1996, Jeff Erickson wrote: > Laura Helen writes: > >Subjects like algebraic geometry, group theory > >at an advanced level, differential geometry, algebraic topology, etc. > >would be considered "graduate-level". > > There's an obvious bias for "pure" "continuous" mathematics here. > Where would combinatorics fit into this newsgroup? Numerical > analysis? Statistics? Discrete geometry? Number theory? The list of graduate-level subjects is not intended to be exhaustive. Yes, the subjects you mentioned would be acceptable -- any *mathematical* discussion at a graduate level would be acceptable. I imagine numerical analysis shades into software engineering, etc. questions -- e.g. "How can I make this algorithm more efficient" or "Will this random number generator be biased if I do it single-precision" -- such questions would have a better place in another newsgroup, but questions of graduate-level mathematical interest would be welcome. Article 71305 of sci.math: Path: news.sri.ucl.ac.be!jussieu.fr!math.ohio-state.edu!uwm.edu!vixen.cso.uiuc.edu!chi-news.cic.net!news-w.ans.net!newsfeeds.ans.net!newsjunkie.ans.net!newsfeeds.ans.net!newstf01.news.aol.com!newsbf02.news.aol.com!not-for-mail From: jeffleader@aol.com (Jeffleader) Newsgroups: sci.math Subject: Num. Analysis is math.! (was Re: Sci.math.grad -- preliminary proposal) Date: 10 Mar 1996 18:38:37 -0500 Organization: America Online, Inc. (1-800-827-6364) Lines: 20 Sender: root@newsbf02.news.aol.com Message-ID: <4hvp5t$jbb@newsbf02.news.aol.com> References: Reply-To: jeffleader@aol.com (Jeffleader) NNTP-Posting-Host: newsbf02.mail.aol.com >I imagine numerical analysis shades into software >engineering, etc. questions -- e.g. "How can I make this algorithm >more efficient" Numerical methods shades into software eng., whatever that term might mean. Numerical analysis has plenty of analysis in it. Making the DFT more efficient leads to the FFT, a nontrivial bit of math. The theory of convergence and stability of numerical methods and of error estimation, for example, involve "hard" math. Finite differences leads directly to Kreiss matrix theorem, etc. A surprising number of big-name numerical analysts knew little about programming etc.--they are/were doing analysis. While an undergraduate level finite differences/elements for undergraduate engineers (like I must give next quarter) might have a large programming emphasis as a way of getting two things taught for the price of one, in all my graduate level num. analysis training we did no programming--we were doing num. *analysis*. It shouldn't be confused with a programmer implementing something and trying to eke out a little more speed or use a little less memory or something. Think Lax Equivalence Thm., Kreiss Matrix Thm., etc., instead. Article 71551 of sci.math: Path: news.sri.ucl.ac.be!jussieu.fr!oleane!plug.news.pipex.net!pipex!tank.news.pipex.net!pipex!news.mathworks.com!newsxfer.itd.umich.edu!jobone!ukma!netnews.wku.edu!netnews.wku.edu!adler Newsgroups: sci.math Subject: Re: Num. Analysis is math.! (was Re: Sci.math.grad -- preliminary proposal) Message-ID: From: adler@pulsar.wku.edu (Allen Adler) Date: 12 Mar 1996 04:11:53 GMT References: <4hvp5t$jbb@newsbf02.news.aol.com> Organization: Western Kentucky University Nntp-Posting-Host: pulsar.cs.wku.edu In-reply-to: jeffleader@aol.com's message of 10 Mar 1996 18:38:37 -0500 Lines: 9 In his sensible posting on numerical analysis, Jeff Leader writes: >Think Lax Equivalence Thm., Kreiss Matrix Thm., etc., instead. Perhaps someone would be kind enough to state these and other main theorems of numerical analysis. Allan Adler adler@pulsar.cs.wku.edu