EL VERDADERO PROBLEMA del Climaquiddick (me gusta más inspirarme el mote que le ha puesto Rand Simberg que llamarlo Climagate; al fin y al cabo en el Watergate la prensa estaba interesada en destapar todo al máximo, mientras en en Chappaquiddick hicieron todos los esfuerzos posibles por taparlo, como ahora) es, más que los emails incriminatorios, lo que trasluce el código Fortran del servidor de la CRU de la universidad de East Anglia:

As the leaked messages, and especially the HARRY_READ_ME.txt file, found their way around technical circles, two things happened: first, programmers unaffiliated with East Anglia started taking a close look at the quality of the CRU's code, and second, they began to feel sympathetic for anyone who had to spend three years (including working weekends) trying to make sense of code that appeared to be undocumented and buggy, while representing the core of CRU's climate model.

One programmer highlighted the error of relying on computer code that, if it generates an error message, continues as if nothing untoward ever occurred. Another debugged the code by pointing out why the output of a calculation that should always generate a positive number was incorrectly generating a negative one. A third concluded: "I feel for this guy. He's obviously spent years trying to get data from undocumented and completely messy sources."

Programmer-written comments inserted into CRU's Fortran code have drawn fire as well. The file briffa_sep98_d.pro says: "Apply a VERY ARTIFICAL correction for decline!!" and "APPLY ARTIFICIAL CORRECTION." Another, quantify_tsdcal.pro, says: "Low pass filtering at century and longer time scales never gets rid of the trend - so eventually I start to scale down the 120-yr low pass time series to mimic the effect of removing/adding longer time scales!"

Leed el artículo completo, porque además es un muy buen resumen del follón, y de cómo se pudo filtrar todo.

Como dice acertadamente Megan McArdle,

The emails seem to describe a model which frequently breaks, and being constantly "tweaked" with manual interventions of dubious quality in order to make them fit the historical data.  These stories suggest that the model, and the past manual interventions, are so poorly documented that CRU cannot now replicate its own past findings.

That is a big problem.  The IPCC report, which is the most widely relied upon in policy circles, uses this model to estimate the costs of global warming.  If those costs are unreliable, then any cost-benefit analysis is totally worthless.

ACTUALIZACIÓN. Una puntualización para informáticos; por lo visto no está claro que el código informático sea Fortran, como escribía más arriba basándome en el artículo de Declan McCullagh, me comentan que podría ser IDL. Como yo no soy informático no tengo la más remota idea, y en realidad no es demasiado relevante, pero ahí queda la aclaración.