[UPDATE: Keith Kloor is all over this, and has reaction at Real Climate to Curry's comments. Check it out here.]
From the comments. here are some interesting and provocative comments from Judy Curry, a climate scientist at Georgia Tech.
The primary frustration with these investigations is that they are dancing around the principal issue that people care about: the IPCC and its implications for policy. Focusing only on CRU activities (which was the charge of the Oxbourgh panel) is of interest mainly to UEA and possibly the politics of UK research funding (it will be interesting to see if the U.S. DOE sends any more $$ to CRU). Given their selection of CRU research publications to investigate (see Bishop Hill), the Oxbourgh investigation has little credibility in my opinion. However, I still think it unlikely that actual scientific malfeasance is present in any of these papers: there is no malfeasance associated with sloppy record keeping, making shaky assumptions, and using inappropriate statistical methods in a published scientific journal article.
The corruptions of the IPCC process, and the question of corruption (or at least inappropriate torquing) of the actual science by the IPCC process, is the key issue. The assessment process should filter out erroneous papers and provide a broader assessment of uncertainty; instead, we have seen evidence of IPCC lead authors pushing their own research results and writing papers to support an established narrative. I don't see much hope for improving the IPCC process under its current leadership.
The historical temperature record and the paleoclimate record over the last millennium are important in many many aspects of climate research and in the communication of climate change to the public; both of these data sets are at the heart of the CRU email controversy. In my opinion, there needs to be a new independent effort to produce a global historical surface temperature dataset that is transparent and that includes expertise in statistics and computational science. Once "best" methods have been developed and assessed for assembling such a dataset including uncertainty estimates, a paleoclimate reconstruction should be attempted (regional, hemispheric, and possibly global) with the appropriate uncertainty estimates. The public has lost confidence in the data sets produced by CRU, NASA, Penn State, etc. While such an independent effort may confirm the previous analysies, it is very likely that improvements will be made and more credible uncertainty estimates can be determined. And the possibility remains that there are significant problems with these datasets; this simply needs to be sorted out. Unfortunately, the who and how of actually sorting all this out is not obvious. Some efforts are underway in the blogosphere to examine the historical land surface data (e.g. such as GHCN), but even the GHCN data base has numerous inadequacies. Addressing the issues associated with the historical and paleo temperature records should be paramount.