home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.edu:1667 comp.lang.fortran:3564 comp.lang.misc:3084 comp.arch:9427 sci.math:11489
- Newsgroups: comp.edu,comp.lang.fortran,comp.lang.misc,comp.arch,sci.math
- Path: sparky!uunet!europa.asd.contel.com!darwin.sura.net!zaphod.mps.ohio-state.edu!caen!hellgate.utah.edu!lanl!cochiti.lanl.gov!jlg
- From: jlg@cochiti.lanl.gov (Jim Giles)
- Subject: Re: Scientists as Programmers (was Re: Small Language Wanted)
- Message-ID: <1992Sep15.165612.7661@newshost.lanl.gov>
- Sender: news@newshost.lanl.gov
- Organization: Los Alamos National Laboratory
- References: <1992Aug31.133811.3626@crd.ge.com> <1992Aug31.135937.5801@dcc.uchile.cl> <1992Sep1.173636.6387@nntpd.lkg.dec.com> <1992Sep10.165629.28124@searchtech.com> <1992Sep14.141559.601@crd.ge.com>
- Date: Tue, 15 Sep 1992 16:56:12 GMT
- Lines: 26
-
- In article <1992Sep14.141559.601@crd.ge.com>, davidsen@ariel.crd.GE.COM (william E Davidsen) writes:
- |> In article <1992Sep10.165629.28124@searchtech.com>, johnb@searchtech.com (John Baldwin) writes:
- |>
- |> | Absolutely. I've seen scientific programs written by self-taught
- |> | scientists which could result in floating point results where precision
- |> | could be destroyed in 1-3 operations. They couldn't figure out WHY they
- |> | would occaisionally get screwy results from some data, and not from
- |> | others. The blame was placed on a compiler bug. Hrrumph.
- |>
- |> I think this is a good example of why a programmer is needed to write
- |> good programs. [...]
-
- Yes, but that still leaves the problem of where to find a programmer.
- The other discussions on this group are revealing the fact the CS
- departments do not (at least, not universally) regard the training
- of *programmers* to be their purpose. Indeed, most recent CS graduates
- I've met *also* don't know anything about floating point arithmetic.
-
- |> [...] With all their training in {physics or whatever} most
- |> scientists don't get a good foundation in numerical analysis. [...]
-
- Since the computing "science" departments don't provide such a foundation
- either, where do we turn?
-
- --
- J. Giles
-