Cheers. No doubt there's additional nuance I've missed but I'm fairly certain he's directionally correct. And, if he is, we face some dire consequences as a nation.
Re the Free Speech Union, that's an interesting one and perhaps points to a broader point. It often feels to me that there can often be an asymmetry of risk faced by participants in some highly charged debates. I know this is a cliche, but there is definitely something to the adage that "conservatives think progressives are stupid, but progressives think conservatives are evil".
So it doesn't surprise me at all that the FSU was founded by somebody from that 'side': If you're debating in an environment when some (I stress some) of the people who may read your writings may actually think you're evil, as opposed to just wrong, it seems rational to invest in some protection?
In any case, I don't know Turver, but I have no reason to believe he's making this stuff up. He seems pretty rational to me, and does share his working. I'd urge you to remain open minded to the (scary) possibility he's right.
He explains on his website where he gets his data from. He gets it from The Low Carbon Contracts Company... y'know: the firm who is the actual counterparty to the CFDs and so should probably know the actual sums of cash being moved - and in which direction.
And note: even when gas is more expensive than the CFDs, the huge fixed and/or policy costs (network build-out, capacity market, curtailment, etc) are devastating.
The story would be completely different if wind farms were actually cheap to build and run... the problem is they're just not.
I wish it were not so... it would be great if we had a path to being free of dependence on hydrocarbons. But in a battle between wishful thinking and physical and economic reality, reality usually wins.
So we're faced with a choice as a nation: continue to pour tens of billions of pounds down this drain... or call time on the experiment and free up all that money for something productive?
The amount wind farms in the UK have contributed back over the last ten years is a rounding error compared to how much they have received. It's not even close: https://x.com/7Kiwi/status/2031657347433603581
And the scary thing: the wind farms aren't even making that much money! Some projects have been cancelled and others had to re-bid in subsequent auctions to get a higher CFD price than they originally received because they couldn't make the economics work. Worse, there are reasons to believe they're not even fully provisioning for their end-of-life decommissioning costs.
The UK's energy policy is unbelievably destructive :(
It's a good question. To be honest, I'm still trying to get my head around how the UK electricity market works. Its complexity is definitely a big part of why so many reasonable people can end up disagreeing so vehemently... vanishingly few people understand how the whole thing works (and pretty much none of those who do are listened to by the politicians...)
Your question is good for another reason: you say "price" without qualifying whether you mean wholesale or retail (and, if retail, what individual households pay or what is experienced by industry). A lot of commentators and politicians routinely conflate the concepts to serve their own agendas in order to confuse non-experts.
If one looks first at the wholesale price, you're right that - in general - one would expect it to 'spike' when the gas prices shoot up. But on days when wind is dominant this has a minimal effect on retail prices, because the extra money paid to the wind farms (everybody gets the clearing price) is exactly offset by a reduction in the CfD payment. To repeat: consumers pay the same (high) price for most wind-generated electricity irrespective of the gas price.
So the interesting question, I think, is: what happens on days when the wind isn't blowing and gas generation is dominant? And here's the thing: if the price for gas-generated electricity (with carbon tax to account for the climate externality) is below the CfD strike prices, we're still ahead, even if it has spiked above its average. And because the CfD strike prices are so eye-wateringly high, this happens far more often than not.
Indeed, it was only for part of 2022 that the wholesale price was above the CfD prices and so the wind farms were paying money in to the system rather than taking out.
Yes - the wholesale price of electricity can sometimes be very low (or negative), particularly when there is a lot of wind. And some tariffs pass that on to consumers. But I don't see how it works at scale.
This is because the wind farms don't get paid the wholesale price. They get paid their guaranteed, index-linked CFD strike price. This means that, for every £1/MWh drop in the wholesale price, they get an exactly matching extra £1/MWh to top them back up to their strike price. They can bid a low price into the market safe in the knowledge they'll get paid their CFD price.
And that top-up has to be paid by somebody: either other bill payers, or the taxpayer.
That wouldn't be so bad if the strike prices were low. But they're not. The recent "Allocation Rounds" guaranteed offshore wind farms in excess of GBP100/MWh, index linked for at least 15 years.
To put it in context, these numbers are higher than the wholesale gas price - even with an uplift for carbon externalities - for all but the worst period of the Ukraine invasion a few years back.
But it gets worse: on top of these extremely high fixed prices for wind, we also have to pay for the installation of tens of billions of pounds (if not more) of new grid connections, because the wind farms are nowhere near the centres of demand. This cost is also added to bills.
It doesn't end there. Readers may be aware that the wind doesn't always blow. Which means we need something that's able to spin up or down on demand. In the UK, that means gas. So we have the ridiculous situation of having to pay the gas plants to sit around doing nothing, just so we can call on them at minimal notice when needed. And remember: there can be long periods with basically zero wind or solar (the famous 'dunkelflaute' phenomenon in winter).
This means we need non-wind capacity pretty much equal to peak winter demand, in order to be safe during the week or so some years when there's no wind.
So we're paying to build and maintain TWO generation systems in parallel.
This is why electricity costs in the UK are on an ever-upwards trajectory: all these 'policy costs' are added to the wholesale price, and are a large and growing component of the _retail_ price that most consumers pay.
Depressingly, 'storage' doesn't fix this. Indeed, it's a fun exercise to calculate how much electrical energy is consumed in the peak of winter in the UK over a one- or two-week period and then figure out how much the necessary battery capacity would cost... or, even more fun, how many Welsh and Scottish valleys we'd need to flood to create the pumped-storage capacity. We're talking tens of trillions of pounds.
I fear the UK has, with the best of intentions, made a mistake of generation-defining proportions with its bet on wind :(
Not a "hard" bug but a useful lesson in any case. I worked on a set of stress tests for a major middleware product and came into the office on a Monday morning to check the 72-hour over-weekend runs. We were getting close to release date and things were settling down so I wasn't expecting anything major. Except they'd ALL failed. It took us far longer than I'd care to admit to figure out what had gone wrong - I wasn't working on it non-stop but I definitely remember it taking quite some time. I think it was a colleague who figured it out later that week.
Anyway, what had happened was that our Perl test harness was tracking time elapsed in the 72-hour run as seconds since the Unix epoch, but was comparing them using the lexicographical order operator (lt versus <). Everything worked until the time ticked over from 999,999,999 seconds to 1,000,000,000.
I just looked up those timestamps to check my memory, and I can now see why fixing it wasn't our top priority that week... the 999999999/1000000000 transition happened the weekend before 9/11.
No need to layer on the flattery... I was hand-waver in chief (and like to think I did a good job in bringing people with us and creating the space for you and the team to work without too much distraction and noise), but it was you who actually brought all the pieces together into a coherent platform. There was nothing in your post I disagreed with!
Hi everybody... Richard Brown, CTO of R3, the firm behind Corda, here.
As Mike says, we're somewhat unusual in that there are many live, successful Corda deployments around the world. Mike's comment about how Corda is blockchain-like but not, strictly speaking, a chain of blocks is at the heart of this I think. Here's what I mean:
How many 'introductory' presentations have you been to (or, worse, given) to semi-technical people, where Bitcoin and other public blockchains are 'explained' by describing the components? You probably know the sort of pitch I mean: they laboriously build up the concepts - transactions, signatures, hashes, blocks, chains of blocks, mining, etc, etc. Such presentation are usually correct. But they impart almost no intuition. It's little wonder that so many 'business people' come away from them thinking that a blockchain is some sort of mysterious and magical technology.
There's a presentation I like to give where I go the other way. I give a one-line description of the problem Bitcoin solves [1] and then I help the audience 'invent' Bitcoin for themselves from first principles. There's an old blog post of mine that gives the rough idea [2]. The point is that successful architectures solve well-stated problems. Cargo culting never works.
Where I think many corporate deployments of blockchain went wrong was that they saw the huge enthusiasm for 'blockchain tech', had a vague intuition that 'inter-firm' or 'market-level' problems were ripe to be solved, but they never fully internalised that Bitcoin's (or Ethereum's) architecture isn't some sort of inviolable, handed-down-from-high blueprint... it's merely a very elegant engineering solution to a well-stated 'business problem'.
Yet many 'enterprise blockchain' platforms seemed to begin with the architecture of a public blockchain, and then tweaked it to make it palatable to businesses (eg engineering cumbersome privacy solutions on top to work around the inherently broadcast nature of the public chains). This always felt a bit weird to me.
With Corda, we were fortunate to have been given the time and space by our backers to write down our equivalent of the Bitcoin problem statement, and then to engineer a solution to that problem. Yes - of course, we knew we were in the 'inter-firm business process' space, and we knew the problem we were trying to solve was in some way 'blockchainy' But we did try really quite hard to write down the problem statement [3] and then go forward from there, rather than starting with a pre-existing architecture and then modifying it.
Yes - as Mike says, architecturally it looks a lot like Bitcoin (eg it has an unspent-transaction output data model). But it also has a bunch of other things that, to this day, no other Blockchain has... eg the Flow Framework that allows decentralised inter-firm workflows to be modelled (think temporal.io but without any centralised infrastructure). And, like Mike says, Corda passes data point-to-point and confirms each transaction one at a time - no blocks, no broadcast.
You would not believe HOW MUCH GRIEF we got from the blockchain community for that design in the early days. Yet - several years on, it seems to be working.
[1] I claim that the 'requirement' for which Bitcoin is the solution is: "build me a system of un-censorable digital cash."
[3] Our problem statement? "Build me a platform that enables multiple firms to record and manage the lifecycle of the business contracts they have with each other, minimising the need for any new third parties." At least, that's what I thought I was building. Mike might disagree, however... I know he likes the "decentralised database" interpretation of Corda.
Are there any MSR designs that require NO reprocessing (whether online, batch or other)?
I'd love to find actual data but, anecdotally, it seems like a large amount (vast majority?) of the nuclear legacy costs for countries like the UK, at least, come from the back-end - ie reprocessing. If you just store used rods (in perpetuity) near the reactor they were used in then the overall legacy footprint is pretty modest. Indeed, I think that's what the UK does now - our reprocessing facilities are now in decommissioning mode and the US did the same a long time ago.
Yes - part of the decision to stop reprocessing was proliferation risk. But I think it's also because reprocessing is so insanely messy and so easy to get wrong.
This is because if you 'reprocess' fuel, the waste problem just balloons... the rods have to be chopped up, dissolved in nitric acid, taken through a complex chemical process and you end up with vast amounts of liquid waste, various bits of undissolved gunk, a fiendishly difficult-to-decommission reprocessing facility and all the rest. Reprocessing plants are some of the most complicated chemical plants in the world... and when they go wrong (eg the UK's Thorp leak) they're almost impossible to repair owing to the radioactivity.
In the past, the purpose of the early reactors was to generate plutonium for weapons and so reprocessing was, in reality, the key activity, with the reactors just the tedious thing you had to build to provide feedstuff for this extraction process.
But if we don't want any new plutonium then there's no need to reprocess and the waste problem just becomes insanely easier.
To see what I mean, google the history of the UK's Sellafield (specifically the B.205 and Thorp plants) or Russia's Mayak or France's La Hague. So many leaks and accidents, all totally unnecessary if they hadn't been trying to reprocess the fuel. The idea of taking something small and stable (a rod) and turning it into a dangerous liquid and then trying to run it through a fiendishly complex chemical plant just seems nuts on its face.
Hence my question about MSRs... can you build one that doesn't require any of this tricky chemical engineering, whether 'online' or otherwise? If so, great. If not, why isn't this whole avenue just shut down as DOA?
Re the Free Speech Union, that's an interesting one and perhaps points to a broader point. It often feels to me that there can often be an asymmetry of risk faced by participants in some highly charged debates. I know this is a cliche, but there is definitely something to the adage that "conservatives think progressives are stupid, but progressives think conservatives are evil".
So it doesn't surprise me at all that the FSU was founded by somebody from that 'side': If you're debating in an environment when some (I stress some) of the people who may read your writings may actually think you're evil, as opposed to just wrong, it seems rational to invest in some protection?
In any case, I don't know Turver, but I have no reason to believe he's making this stuff up. He seems pretty rational to me, and does share his working. I'd urge you to remain open minded to the (scary) possibility he's right.