[repsilat]

I figure it doesn't make sense to commit to it if they can do it without committing, but not a bad thing to give it a go. I figure

- most ideas are bad,

- judgement is a crapshoot but improves the odds up to 50/50,

- trying and either rejecting or committing (or going bankrupt) gets the odds of the better solution prevailing in the market up to 60-odd percent.

Plenty of confounding variables, lots of opportunity to get unlucky and draw the wrong conclusions (or just have your priors overrule weak evidence against them), but it's science and we haven't found anything better yet.