This is a variant associated with the Scully and Druhl signal-idler photon delayed choice experiment, as described and discussed in Brian Greene's Fabric of the Cosmos. The commentary notes the "delay" may be configured to happen a very (years) long time after the photons pass the two slit/splitter. It is stressed that the future measurements do not change the pathway decision "today" but clearly there is a linkage in some "spooky" way. My thought is that it should be possible to set up the "eraser" in such a way that the which-path information contained in the "tagged" idler photon is obscured when it hits the detector "tomorrow" and is recorded, but that in some years time when some technology enhancement is enabled, the recorded detector information can subsequently extract the which-path information (a bit like dna analysis being used on cold cases). The outcome being that when the initial observation/measurement is made, the interference pattern should be restored, but at a subsequent time it should be possible to identify the path taken by an individual photon.