On Tuesday, July 21, 2015 at 5:50:39 PM UTC+12, Paul B wrote:
"Our new analysis suggests that the apparent hiatus may have been
largely the result of limitations in past datasets"

Yes, if one set of datasets does not support the predetermined view, no problem, just get new datasets, or "better" analysis, problem is fixed .

My thoughts exactly.

It is quite apparent that these people DO NOT WANT the alleged problem to prove to be a non problem. They cheer every time some new evidence can be construed to suggest that we're all going to fry, and seem quite upset every time some new evidence suggests that it's all ok, actually.