> It's not how many people have been harmed, it's how many people could be harmed. Really? So we should take into account scenarios like 1. Solar installer fal

> It's not how many people have been harmed, it's how many people could be harmed. Really? So we should take into account scenarios like 1. Solar installer falls from roof, hits car, dies 2. Car didn't have handbrake on, starts to roll down the driveway, rolling over the toddler playing in the yard, killing her too 3. Rolls into the street, causing a school bus full of children to swerve and hit a LNG tanker, causing a massive explosion killing everybody onboard. Of course, in reality, we need to take into account realistic failure pathways and plan accordingly. Per such analysis, both solar and nuclear are both very safe.