Scientific models and doom mongers
Short but deep essay on errors in Scientific Software and Models used over on
Crumb Trail
bq.
Random Answers
We hear a lot of FUD as well as legitimate complaint about buggy software. It wastes our time and resources, offends our aesthetic senses and in some applications can even be life threatening. But it seems that the problems with commercial software are trivial compared to the software on which scientific research is based. See the most recent posts at
Notional Slurry for an inside look.
Excerpting two more paragraphs:
bq. This is important. Much of the alarmism spouted by various doom advocacy groups is heavily dependent on models. Their record over the past few decades has been laughable as time and again their projections and predictions based on faulty analysis were demonstrated to be nonsense. Knowing that the quality of model implementation is as poor as the analyses on which they are based underscores the importance of keeping science at arms length from policy and exercising due skepticism about the hysterical claims of activists.
bq. Those with authoritarian leanings and a need to believe are uncomfortable with uncertainty and unable to formulate policies that give it proper consideration. The media is filled with hysterical screeds denouncing one government or another for ignoring science in policy making, but this can be a very good thing. Scientists are wrong so often that it's sensible to let their findings age and mellow before use. We might take a page from scientific institutions which often wait decades before awarding honors and prizes for discoveries since they know that most will be shown to be flawed when thoroughly examined.
Very well put - the entire essay is five paragraphs - worth your time reading if you are at all interested in Science and Public Policy making.
Posted by DaveH at February 20, 2004 4:37 PM