Retconning the data? That's highly unpopular among a vocal minority. Typically these sorts of things show the raw data and allow people to extrapolate 'real' in

Retconning the data? That's highly unpopular among a vocal minority. Typically these sorts of things show the raw data and allow people to extrapolate 'real' infection rates etc for themselves. Changing all data from before there was a change in the measurement standard would be putting extrapolation right into the data feed. I think the responsible thing to do might be to show error bars, then it would be more intuitive what happened.