Good questions about climate change

From Jerry Pournelle. [Link]

Please tell me how to determine (1) the temperature of the Earth in 1895 -- what operations do I go through to generate that figure, (2) the same for the year 2000, and (3) exactly how to decide what weights to give the ground temperature at Santa Monica airport, undersea temperatures at various latitudes, stratospheric temperatures, and other measures to generate this single figure of merit.
Does anyone know how this is decided?

Continuing: once we determine how to calculate that figure of merit, can well all agree it has been done properly? That is take the original data that generated the 1895 temperature estimate and recalculate that?
Given that the figures of merit are calculated properly, what was the warmest year of the 1901 - 2000 period? Were any adjustments made to the method of determining the single Earth temperature or were the same operations used for each given year? If the generation of the global temperature was adjusted or refined, how, and why?
These are elementary operations. The formulae can be published in a single book, but I haven't seen that book. The data can't be much larger than a couple of gigabytes. The multiple factor equation can't be that large -- a thousand weighted factors? I presume it is linear, meaning that it's a few seconds calculation, and the program to do it can't be much more than a few hundred lines of FORTRAN. The climate models may be incredibly complex -- more so than the Model of Doom that was so popular in the 1980's -- but an explication of their flow charts and some measure of the sensitivity of the outcome to given input elements can't be that difficult. I have never seen any such thing or even a reference to one, but I may not have spent sufficient time on it. Still, I can't find students who have been told where to find such an explication.
We have for years been told to pay attention to the IPCC political summaries which are based on underlying science that is conveniently left out of the reports to political leadership. But the underlying science doesn’t seem so transparent and some prominent scientists have quit because of the ways the summaries are generated.
The point here is that we are dealing with decisions that allocate trillions of dollars and have enormous effects on global economies. We do so with what amounts to suppression of actual debate on the science involved. That was the importance of Climategate: that the peer review system itself is being manipulated. That's no surprise, since it has been happening for generations.
Before we spend trillions on corrections, it seems reasonable that we should spend tens of millions on the best tests of the underlying hypotheses and data that generate the recommended actions. We ought to try to find competent people who are unbelievers, deniers, and supply them with resources to allow them to attack the consensus and ask embarrassing questions. The whole operation wouldn't cost more than was spent on the East Anglia consortium, and would sure go a long way toward relief of doubts.