The problem is that AGW the researchers like Michael Mann at Penn State are not releasing their source code or raw data. So there is no way to verify their computer models independently.
Back in late 1980s when I was a CS undergrad at UMN and later a CS grad student at UIUC, I worked quite bit with researchers to fix up their Fortran simulations, doing code optimization for models run on Cray supercomputers and TMC Connection Machines at the MN Supercomputer Center and at NCSA.
In those days all the big federal granting agencies (National Science Foundation, National Institutes of Health, etc) required that any computer source code created or funded using federal research money must be made available to any US citizen on request. I spent plenty of time at the copy machine killing trees and spinning tapes for these requests. I tracked my time for billing purposes.
Some time after I left (not sure when) the federal research agencies quietly changed the rules. The requirement for public disclosure was dropped.
That is the whole problem. Bring back the requirement that, to get federal funding, you must make available your raw data and computer source code to anyone on request (and if they pay reasonable expenses for providing it).
That was the old rule in the 1980s.
It would fix the whole problem.
“Back in late 1980s when I was a CS undergrad at UMN and later a CS grad student at UIUC, I worked quite bit with researchers to fix up their Fortran simulations, doing code optimization for models run on Cray supercomputers and TMC Connection Machines at the MN Supercomputer Center and at NCSA.”
Cool! I did the same thing at the National Center for Atmospheric Research in Boulder in the late 1970’s as a Research Assistant farmed out to NCAR by the CU Computer Science Department when I was getting my MS in CS at CU.