Steve McIntyre's weblog Climate Audit has always been a hotbed of Climate Change Denialism, and he has spent much of the last several years flailing about madly in an effort to discredit the science behind AGW. However, it seems he might have hit onto something this time, having detected an error in the Global Historical Climatology Network's data collection procedures which, when corrected, reorders the the ranking of 10 hottest years recorded in the lower 48 states of the United States such that several years in the 1990s and 2000s fall out of the top ten, and several years from the 1930s come into it.
The new and old top ten lists can be found here, and the numbers for the last 120 years, as corrected by NASA itself, can be found here.
Now, there are lots of mitigating factors to consider. For example, the new adjustment does not seem to affect global average values much (these are calculated separately using results from a number of distinct networks). Nevertheless, grist for the skeptics mill, and material that will get huge play over the course of the next several months.