Michel Caillaud weist zu Recht darauf hin, dass nicht alle Springerbewegungen im Retro der Woche 45/2013 eindeutig sind; er schreibt:

The retroplay doesn’t look unique in the problem posted on your blog :
2.Sh5-g7 Sg7-e6+ 3.Sf4-h5 e5-e4 4.Se6-f4 Sf3-g5 5.Sg5-e6+ Sd4-f3
and also
4.Se6-d4 Sf3-g5 5.Sg5-e6+ Sd4-f3
and also
7.Sc8-e7 Se7-c6+ 8.Sb6-c8
seem possible alternatives.
Of course, the answer to the stipulation is still correct, but the claim of unique retroplay doesn’t stand.