[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On 22/04/2013 07:41, Henry Bremridge wrote: > On Mon, Apr 22, 2013 at 07:11:23AM +0100, tom wrote: >> Given that most economists seem to think the economy can not be >> modelled accurately in the first place I cant imagine them worrying >> about coding minutiae. > > The importance is making sure that careless errors are found > I think the importance is knowing what the automatic formulae do , and getting your reasoning (nothing to do with spreadsheet) right. Different weighing to summations will always be wrong whither calculated on a spreadsheet or abacus of by logarithms. Plus users did not understand how spreadsheet formulae worked. (Roundings etc) For example 'average' has at least 7 different interpretations. Politicians (learned in PPE) usually have little or no mathematical training and thus cannot question results. A better example is how the difference in a few cents in bill on many thousands of dollars from telephone company against logs of use came to find outside (Chinese) penetration of USA government computers by hopping (enter at university due security error then immediately transfer out to government computer) via an academic establishment. Dismissed as rounding errors by all financial folk and authority in University until a mathematician showed the two computers (university and telephone company) used same logic and must get same answer. Then a great problem. While I followed the investigation, and know of the case , I forget the exact Mathematician's name. Politicians do not question 'executive summaries'. -- Regards Eion MacDonald -- The Mailing List for the Devon & Cornwall LUG http://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/listfaq