Update: See new graphs at the end — no “cherry picking.”
October 13, 2012 ————

I’m not a climate scientist – or any other kind of scientist, for that matter.
However, I can take a set of numbers and dates and create a graph from them in Excel.
The preceding graph represents monthly global temperature variations in °C from January 1997 through August 2012 – almost 16 years.
I produced the graph after seeing this one at the Daily Mail:

The following graph is from the same data source, except it is “decadally smoothed” – “filtered to remove variability on time scales of less than a decade.”

This is the first time that I’ve actually plotted any of the data and I must say that I’m a bit surprised, especially with the second figure.
The dataset is part of the HadCRUT4 time series and comes from the UK Met Office Hadley Centre. It is available here.
__________
October 16, 2012
1850 to 2012 global temperature HadCRUT4 anomaly:
1900 to 2012 global temperature HadCRUT4 anomaly, “decadally smoothed” – “filtered to remove variability on time scales of less than a decade.”: