I always thought the Karl, et al 2015 paper was odd, in that it compared times frames the earlier studies did not.
Possible artifacts of data biases in the recent global surface warming hiatus | Science
The problem with this statement is it contrasts greatly with the other stated periods of the observed warming.
Global warming (GISS)from 1950 to roughly 1978 was almost ZERO, The period that caused the concern has been stated as 1978 to 1998.
To average in 28 years of ZERO increase will flatten the per decade rate of the earlier period.
If we took the 5 year average in 1978 (.006) to the 2000 5 year average (.46) we get a per decade rate of .206 C per decade.
The 5 year average from 2000 to 2015 was .14 C per decade. (Delta .21C over 1.5 decades)
So there is indeed a noticeable change in slope.
Changing the slope of the "before" line by including a longer period of zero warming, just demonstrates that
the long term rate of warming is lower than expected, and not very alarming!