What safety standards do you think nuclear power plants must meet beyond a shadow of a doubt before a restart of their reactors should be considered?

it is not for me to say as i am not a scientist but as a human who must live with nuclear power - BRING IN EXPERTS and GET SERIOUS about safety checks, back-up systems, workers. in other words - transparnecy, checks and balances.

That's very simple : make is that the ground where plant is built is NOT the planet's most frequent earthquake zone. Safety of nuclear in general is one thing, Japan has always been a special case. That always was a bad idea here,

Obviously, full redundancy in cooling systems. Also: make plants subject to regular IAEA inspections by non-Japanese inspectors: maintenance of a fleet of dedicated vehicles and robots to respond to accidents: ban on hiring migrant part-timers and paying them minimum wage. The list is a long one.

"Beyond a shadow of a doubt"? Well that's kindof impossible, but I'd say that they definitely should meet all current safety recommendations before being bought back online. I like the car analogy for this. Companies that own fleets of trucks are required to ensure that their vehicles are in good running condition every two years, and the truck isn't allowed on the road unless it meets ALL the safety standards (tyres, engine, brakes, lights, brake fluid, etc). Why should a LESSER standard be applied to nuclear power stations?

It certainly isn't a case of "the public good", since the government is adamant about applying strict standards to trucks that deliver our food. There's no reason to apply a lesser standard to the power stations that deliver our power.

How about making sure that they aren't on the coastline where a tsunami can crash into them.

Would be a good start.

Make em so good that the Japanese leadership would agree to have the plant right next to the Diet building.

They'd save a fortune on transmission costs too

Well to start how about concrete plans for how they are going to take care of the spent fuel that will continue to be lethal well beyond the end of our civilization. I also recommend that the people in charge of safety standards be forced to watch "Into Eternity." As I first stated: just for a start. There are many many many other things too.

They shouldn't have greedy old men at the helm for starters.

Seriously?!? Wow. Oh geez, I dunno... ALL OF THEM!!!

What is so difficult? Nobody else has problems upholding standards.

Simple, really. Price the cost of nuclear energy to include any and all payments should the plant experience a meltdown. Then, charge customers accordingly. Keep any income over a certain percentage (for profit) in escrow in case of the meltdown. After the plant is safely decommissioned (in 50 plus years), the operator can have a percentage back and the customers can have a percentage back. That should encourage plant operators to follow enough safety standards, perhaps?

And, yes, if the price includes clean up for a meltdown, it might be too expensive to use.

There are not any standards that can cope with undamped fuel rods-if there were then they would have been put in place already.

The question is blatantly hypothetical!

Shut it down and never play with dynamite.

The problem I got with calling in experts.

Which experts, so far I got no answer to that one. Who has experience with a disaster like Chernobyl or Fukushima globally, besides on paper and simulations.

True answer is NO_ONE, they are all playing it as it develops. The IAEA has access to the experts globally and that includes(belief it or not the japanese who build and designed many plants globally).

The USA has extended the life-expectancy of their own plants recently by 10-20yrs, nearly every country including france has had problems(like very recent).

I have worked in IT(highly trained) and when we got stuck and called on the makers and designers and were told to RTFM(READ THE F....ing Manual). And that was from the IBM Black Team. At times we told the makers and designers how to fix their rubbish, as we often evaluated their software on request.

The same seems to be happening here, no-one has the answers.