Japan data confusion mastered

Confusing me, the analysts went to percent change in growth rate, over the period. Normally I am used to growth rate as a percentage from a year ago, or even compound growth rate over long periods. But with a regime change in Japan the percent change from a year ago does not show the effects of short term changes. Hence the switch in units of measure made everything appear out of whack. Units! Always changing and often the GDP report does not specify exactly what units.