It seems to me that the single biggest issue with Climate Science is the lack of replication. Science needs replication (remember cold fusion?) to validate the results.
With global temperature sets being released in the form of "valued added" product and with statistics and simulations being such a significant part of climate science, the only way to replicate results is to release the code and data.
Hence my new credo:
In God We Trust - all others show code and data.