Hacker News new | past | comments | ask | show | jobs | submit | Gwypaas's comments login

Arnold Schwarzenegger hade an obese BMI in competition trim. Outliers does not invalidate a generic measure, they are outliers.

https://upload.wikimedia.org/wikipedia/commons/0/05/Arnold_S...


The argument of "BMI doesnt account for muscles/height" always sucks because of what you said, they're just outliers. Realistically nobody that is TRULY an outlier takes BMI into account in anyway.


Is not this common knowledge? It works for population measures and mostly everyone not exercising at a very high level. In other words, 9x% percent of the population can not excuse their BMI by "muscle mass", it is simply due to a sedentary lifestyle and the BMI is a correct estimate.


> You're imagining an act of god that does precisely so much damage that the difference between catastrophic failure and a non-event is this one cracked pipe. That was, please note, detected and fixed before failure.

Take the severe Forsmark incident in 2006 in Sweden. Many of the "defense in depth" layers had been accidentally removed through freak occurrences and upgrades. Thus loss of cooling became almost a certainty. That is why you test and do not accept half fixes to placate the operators profit margin.

https://en.wikipedia.org/wiki/Forsmark_Nuclear_Power_Plant#J...

> And the downside of that risk coming to fruition is a 0-death to near-0-death crisis (which is much better than whatever this precise external shock will cause).

And a at least $200B bill to cleanup the mess in Fukushimas case. Lets remove the Price Anderson Act so they have to pay the true cost for their risk?

https://en.wikipedia.org/wiki/Price%E2%80%93Anderson_Nuclear...


> Lets remove the Price Anderson Act so they have to pay the true cost for their risk?

Yeah. The costs and benefits should be born by the capital owner. The issue is that if you want to force them to pay for a cost externality that should be balanced out by considering the benefit eternalities. For nuclear power? if the world was fair they'd get a much bigger net subsidy. The risks of nuclear power going critical are far smaller than the benefits from not having to use coal for example.

I forget what a life saved is worth in engineering terms. Something like 1 or 10 million per capita I think. $200 billion in cleanup only needs to save ~200-2,000 lives to be justified.

> Many of the "defense in depth" layers had been accidentally removed through freak occurrences and upgrades.

Things like pipes being cracked, for example? That is the issue here to me, this is part of a system of defences where it is anticipated that some of them won't be working. No one defence being broken should be a crisis.

I'm cool with the idea that they should fix their pipe. I'm not cool with it being treated like a big deal without pretty solid evidence that the deal is big.


> The risks of nuclear power going critical are far smaller than the benefits from not having to use coal for example.

That is a strawman argument. Coal has generally been uneconomical since the advent of combined cycle gas turbines where gas infrastructure exists. Today renewables are vastly undercutting both.

Trying to frame it as a choice between nuclear and coal is only made because nuclear does not stack up against the real competition in 2023.

> I forget what a life saved is worth in engineering terms. Something like 1 or 10 million per capita I think. $200 billion in cleanup only needs to save ~200-2,000 lives to be justified.

Or we just build power generation without those third party risks. I do not understand why you are trying to frame a $200 billion cleanup bill as "nothing to see, move along sheeple!".

> I'm cool with the idea that they should fix their pipe. I'm not cool with it being treated like a big deal without pretty solid evidence that the deal is big.

Almost all nuclear accidents final hole in the Swiss cheese is some sort of loss of cooling. When the backup power fails due to negligent maintenance that is a big deal.

Fukushima led to us building stockpiles of backup generators together with the necessary electric connections allowing us us to fly them in with helicopters and connect them if the primary ones fail. Saying that failure in the primary ones is "fine, nothing to see" is sticking your head in the sand.


Assessing a 1957 Act against the performance from then to today is hardly a strawman argument. Since we're talking about US Nuclear, we're talking about a design, risks and benefits of what is effectively 1973 tech.

If you want to talk about the future, then guessing at $200 billion cleanup bills seems unlikely. The design state of the art has come a long way since the dawn of the nuclear industry. We don't know what a serious failure of a post-Chernobyl design looks like because such a thing has never happened.

> Or we just build power generation without those third party risks. I do not understand why you are trying to frame a $200 billion cleanup bill as "nothing to see, move along sheeple!".

You are, at present, living with a >$200 billion cleanup bill because we didn't commit to Nuclear early enough. Given past performance, the odds of losing more than $200 billion from not enabling Nuclear now are quite high. Net benefits outweighing costs isn't that complicated a stance and I expect you'll understand it if you think about it.

And take note on the way through that the renewables proponents have serially overstated their case for years - Germany appears to be on the path to de-industrialising itself rather than admitting that its renewable program was actually quite expensive. I'd much rather have $200B as a one-off cost rather than the ongoing fiscal disaster of their Energiewende. The cost of renewables has improved a lot since then, but it seems unlikely that the situation on the ground is as lopsided as the raw costs suggest - being able to schedule the generation of power counts for something.

Also, what is with this "sheeple" business? I never said that. And I'm addressing the issue of cleanup cost directly.

> Almost all nuclear accidents final hole in the Swiss cheese is some sort of loss of cooling.

Yeah. If the cooling doesn't fail then it is pretty hard to see how accidents could happen. But the argument isn't that accidents won't happen - accidents do happen. We can only control probabilities and sooner or later everything fails.

The issue is that the situation is a lot like the UK in the 1750s refusing to use coal because of the risk of air pollution. They'd technically be right about the costs, but the upside of cheap power is much, much more important than the downside of things going wrong. Nuclear has a smaller cost than coal and a bigger upside. Pulling the plug on the nuclear industry was and remains foolish policy.

> Fukushima led to us building stockpiles of backup generators together with the necessary electric connections allowing us us to fly them in with helicopters and connect them if the primary ones fail. Saying that failure in the primary ones is "fine, nothing to see" is sticking your head in the sand.

I'm not sticking my head in the sand, I'm saying that a hypothetical failure of one component (which didn't happen and was caught by the inspection program) is acceptable. You just provided a new argument for why. These things have a lot of redundant layers of protection.



Or Dropbox and choose "online only".

For my off site backup I have a computer which syncs everything to local storage. I guess I don't defend against malicious intent or "delete it all" software bugs but that tends to end up as en exercise in insanity.


Storing directly on Dropbox is 5x time slower than on 'normal' filessystem.


Which is why I treat my computer as a thin client unless it is in the Dropbox folder.

For example, every single line of code is simply a scratchpad if I don't sync it to a remote git instance.


https://www.lazard.com/research-insights/2023-levelized-cost...

> The same argument according to which solar panels and such will become cheaper as demand increases (because of economies of scale and technical improvements) also applies to nuclear power plants.

The French nuclear scale-up famously did negative learning by doing. Economics of scale is not something you can assume. This is from before Flamanvile 3 started, that is even more negative learning.

https://www.sciencedirect.com/science/article/abs/pii/S03014...


> Not to mention nuclear is also the safest technology out there so far (read up stats on the deaths per each technology, per megawatt)

Let's remove all laws where the public bear the cost of accidents then? Should be fine since it's the safest!

https://en.wikipedia.org/wiki/Price%E2%80%93Anderson_Nuclear...

That is for the U.S., but equivalent laws exist in all countries with nuclear power plants.


Feel free to do so! I'm not interested in law in US, nor is law my point in general.

Safety of nuclear is confirmed by death stats and bills like the one you mentioned don't prove otherwise. In fact I'd say they encourage lesser safety measures if anything...


With reduced safety features that would not be acceptable in the west. Then sprinkle middle eastern style guest workers and a dictatorship on top. Will not be repeated in the west.


How do you know the “safety features” will be substandard?

I worked at Jebel Ali power plant(worlds largest gas plant) just outside Dubai for two years installing and commissioning high voltage switchgear.

The Emiratis(and their army of Indian/Bangladeshi/Pakistani/misc./ engineers and laborers) truly knows how to build large scale industries. I doubt they have taken any shortcut whatsoever.

In my experience, they require everything to be 110% perfect.


> “Nuclear reactor design has evolved, but key additional safety features have not been included at Barakah, with the chief executive of Areva, the French nuclear cooperation, comparing the Barakah reactor design to, quote, ‘a car without airbags and seatbelts,’” he said. “So the Barakah reactor design may prove inadequate defence against significant radiation release under what’s known as ‘fault conditions’; in other words, an accidental or deliberate airplane crash or military attack.

> “And what’s particularly worrying is the lack of a core catcher, which in the event of a failure of the emergency reactor core systems, would retain the nuclear fuel once it breached the reactor pressure vessel. On top of that, concrete cracking in all four reactor containment buildings hasn’t helped, nor has installation of faulty pilot-operated safety relief valves.”

https://www.power-technology.com/features/divided-opinion-in...


So a chief executive from a competitor is unhappy with the safety features of the reactor.

Do you have a less biased source?


Nonsense these are highly safe plants from South Korea. They are very high safty.


Is there any public information about the financing?

The recent nuclear plants constructed in the west requires something like $5-15B above the market price in subsidies, of course depending on size. Would be interesting to know how they intend to solve that.


A lot of things are weird about Poland finances. They buy very expensive military hardware in huge quantities, and now a nuclear plant? Were does the money come from? And were will the money for all the necessary maintenance come from in the future?



No, EU doesnt pay for their military purchases.


EU paid for some/most of old gear we send to Ukraine. Poland was the first to send tanks while Scholz was busy scaremongering about WW3 on TV and deleting armor from list delivered to UKR, over three hundred PT-91 modernized ex soviet ones + Leopard 2A4s instead of promises of some in 2024. We send jets before public declarations, we send ~100 modern AHS Krab artillery pieces, >50 Rak 120mm arty, 200 modern Rosomak APCs, 30 Oncilla APCs, 12 Mi-24 helicopters, some S-60 old soviet towed AA and 50 deathtrap BMP-1s (still better than civilian trucks). We didnt dig old decommissioned stuff out of museums or deep storage, it all came straight from active use in military units. Purchases you see announced are backfill for that equipment, we are out of stuff to send.


>while Scholz was busy scaremongering

Scholz is a meme.


You might be surprised by the net inflow vs. outflow of funds, or where these EU funds go, not just a narrow EU statistic.


I am quite confident that by any measures Poland is a net beneficiary of EU; and as a citizen I am quite happy about that situation.


> Were does the money come from?

Inflation...


PGE will finance it on-balance sheet, the way they would a coal plant. Most likely there will be US export finance for the elements coming from the US but most spend on big nuclear plants is local.

>The recent nuclear plants constructed in the west requires something like $5-15B above the market price in subsidies, of course depending on size, EPR at 1600 MWe or AP1000 at 1000 MWe. Would be interesting to know how they intend to solve that.

I'm not really clear what this means? All power plants require financing for their construction cost which is pretty definitionally the market cost of construction. If you mean how this will will interface with the Polish electricity wholesale market, well that market is already pretty dominated by a few large vertically integrated players settling trades with themselves. Presumably PGE will buy the electricity and treat it the same as their other electricity, they may receive a subsidy per MWh as well or maybe just avoiding the ETS costs of their current coal emissions will be enough to make the economics stack-up.


> I'm not really clear what this means? All power plants require financing for their construction cost which is pretty definitionally the market cost of construction.

It is hard to finance a plant when you will on average sell the power at a loss. The money has to come from somewhere.


The Polish wholesale power price is above even what nuclear power costs to deliver in higher-cost UK and US markets on an LCOE basis (not that LCOE is a great way of comparing power costs but there it is).


You have to differentiate between paid off nuclear plants nearing end of life and new built ones.

Take for example Hinkley Point C, that costs €0.15 per kWh for the consumers. How does that compare?

https://en.m.wikipedia.org/wiki/Hinkley_Point_C_nuclear_powe...


Of course nothing is forever. But why can't we celebrate the win that BEVs are?


_hypx is a hydrogen proponent and believes that is a superior solution. "HypX" is also the name of a hydrogen car company.

https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...


That is a pure coincidence.


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: