Path: utzoo!utgpu!watserv1!watmath!att!att!linac!pacific.mps.ohio-state.edu!zaphod.mps.ohio-state.edu!sdd.hp.com!caen!umich!bushido!dbc From: dbc@bushido.uucp (Dave Caswell) Newsgroups: comp.lang.misc Subject: Re: C's sins of commission Message-ID: <1990Nov18.031503.2728@bushido.uucp> Date: 18 Nov 90 03:15:03 GMT References: <27376@megaron.cs.arizona.edu> <-M.6315@xds13.ferranti.com> Organization: Bushido Systems of Ann Arbor, Michigan. Lines: 29 .As to this I have good news! DBMSes and 4GLs are often (anectodal but .frequent evidence) impressively *faster* than any ad-hoc program that .does the same. NOTE: *the same*, not similar but much much simpler. . .If you are manipulating large masses of data in complicated ways, which .is most people want to do, a well written DBMS is going to be way ahead .of any well written lower level solution. Notice that you *can* for a .very important problem outrun a well written DBMS with diabolically well .written low level code, but this is a rare case. . .The same applies for example to sort utilities -- normally it takes a .lot of effort to beat a well written generic sort utility on any system .I know of. This sure doesn't my experience. My experience is that custom written code can take a problem that takes 10-20hrs. to solve and change it into a 10-20 minute problem. In other words the custom written code is what allows the solution to actually be used. I'm not saying that Oracle or whatever isn't a very useful tool just that there is a whole host of problems that it can't solve. Many of these problems are relatively easy to hand-code using well known algorithms. (well known means published in CACM or something similar). -- David Caswell dbc%bushido.uucp@umich.edu