That why we use turbo codes, or at least interleaves. The problem with RS codes is that to optimal decode them requires quite a bit of hardware/cpu cycles.

Err no. It is already very clear where liability lies already covered buy existing laws. Equipment failure that costs lives and causes damages are not new. Neither is software failures.

You mean like all the RBMK reactors that are all decomissioned after the very serious flaws in its design where exposed after Chernobyl? Oh wait they still operate almost all of them still.

Russia won't even publish full accident reports and has a awful history with safety. Russia being the only place that runs them is not a point in favor. Quite the opposite.

Oh and liquid salt reactors come in Fluoride or Chloride flavors. Chloride salts are soluble. At elevated (aka operating) temperatures Fluoride salts react with water.

Me.

The thing is they could be really shit and still be better. But, perhaps for the best, autodrives will be held to quite a high standard.

Cars already have mechanical failures that kill people. There more than enough legal examples of how liability works in cases like this.

So when the time comes, get a classic car and get it towed to the race track. Just because you want to here the roar of the motor does not mean the rest of us should put up with the shoddy, slow to react and down right poor wet computer driver.

Not anymore.

A better question is, should we have non override autopilots? Once good enough i say yes. But don't be fooled most planes don't have as much automation as people think.

I was ridding motorcycles over the farm on my own at around 5 years old. Some of my friends started younger. People are just far too risk adverse these days. Not only are you not decreasing risk. But getting no valid tradeoff either. Here in much of the EU very young children get the public transport on their own all the time.

Why is it stupid. Getting rid of the stupid human seems a plus. Our great intuition and abilities kills far more than it saves for such repetitive mechanical tasks.

There are already plenty of similar such things. Many subways and trains around the world have no driver.

The thing is that the volume reduction is about 65x. Quite a lot. In fact with correct on site processing and neutron spectrum control, and sub critialicaly in the right places. A 1GW plant needs a fairly small room to store all active waste for the required times. That is after about 100-200 years, the first stuff you put in there is now safe and can be taken out.

Such a nuclear plant however would be a major R & D project and there is nothing to say it would be cost effective.

I think you need a new tin foil hat, the current one you have is not working.

No it wasn't. Its a lot more complicated and more difficult that plain PWR and BWR. Plain and simple. Also solid fuel designs keep the radiation much more localized which makes maintenance much cheaper. Thorium salt reactors are a very long way from ready or even so much as demonstrated. Sure there was a small one, that *didn't* breed, that *didn't* do the in situ processing, that was tiny and only *suggested* materials to fix the corrosion problems. It did not demonstrate any of that. On top of all that is was only 10MW.

Every liquid sodium reactor built to date has suffered periodic sodium fires. Its not as ready as you think. Also you are leaving out the fact that fast reactors are much harder to control and even more so, as in impossible, with too much actinides in the fuel. Sub critical assembly's with ADS was suggested for this reason.