The olympic marathon swimming was indeed mean. In any case, it's up to the sports organisations to determine what the rules are and what the rules should be. I don't say it's wrong what we're doing in rowing. In fact, using a photofinish with a minimum resolution of 1/100th of a second is an easy and pragmatic solution.

But it's not necessarily the way things have to be forever. My main concern is that it pretends to offer a precise measurement of relative performance when in fact it cannot because we can't guarantee a similarly precise alignment at the start. Or at least the rules don't ask for it (accuracy of alignment + precision of its measurement is not prescribed in the in the FISA rules of racing).