> Starting a new process for each request doesn't feel right.
I think there is a misunderstanding of how stdio works. The process can be long running and receive requests via stdio at any time. No need to start one for each request.
Sure, but that is not what I'm seeing in FastMCP proxy mode, and it can only talk with the parent process. You make a good point though, stdio is similar to tcp sockets and cross-platform without namespace issues, but can only support one client per process. I could use socat if I want it to talk sockets.
And also the ruinous planning regulations in the UK, which thankfully the current Labour government have signalled they're going to reform (which hopefully means gutting a lot of them).
This sort of numerical deception is endemic in nuclear advocacy dicsussions. I don't know if it's intentional, the best explanation I have is that somebody proposes an (unrealistic) number as a goal, and then it's accepted by advocates as the reality. But if reality was known, the person probably would not be an advocate.
I have not yet met an advocate that will talk about the real, existing, challenges to nuclear the same way that renewable energy advocates talk about the challenges. Go and listen to a podcast on renewables, and it's all about what's not working, what's working, where the opportunities are, and what needs to be changed with the tech. Go listen to a nuclear energy podcast and all you get is wishful thinking, rose-colored glasses, and papering over the real problems that face the real projects.
Edit: for example of the podcasts, here's Jigar Shah on Decouple https://www.youtube.com/watch?v=HwN1MCtBkVk giving the hosts a dose of reality. And Shah is far more optimistic than nearly any realistic person on nuclear, the Liftoff plan (https://liftoff.energy.gov/advanced-nuclear-2/) that his department produced starts from an unrealistic current position and then assumes massive gains somehow to get to a competitive position. It's rose-colored glasses in itself, but at least it shows a plan of how to make nuclear realistic.
My personal theory is that a lot of nuclear advocates are pro-nuclear first (possibly to be contrarian, or because they think the tech is cool), then look for favorable facts/data second.
Meaning the advocacy is not driven by best-effort current estimates of efficiency/cost/benefits-- its the other way around, leading to such a high number of overoptimistic takes on it.
There are lots of wind and solar advocates completely glossing over the difference between nameplate generation capacity (maximum capacity) and intermittency and the value of an intermittent source in the market (marginal rate payment schemes vastly overvalue intermittent sources). That doesn't make it ok for nuclear advocates to do the same, but your portrayal of nuclear advocates as uniquely biased is inaccurate.
I agree with you that Jigar Shah knows what he's talking about and is worth listening to, but disagree that he is some magical thinker wishful optimist about nuclear in particular. He would say that basically, by the time you include the cost of additional gas generation (which is like, fully supply-chain constrained through 2030+) and battery storage, everything costs more than $100/MWh and nuclear looks competitive, even at Vogtle 3/4 prices and especially with some experience curve effects.
At the same time storage is absolutely plummeting in price with the latest auctions in China landing on $63/kWh serviced for 20 years.
Then we have comprehensive grid modeling:
See the recent study on Denmark which found that nuclear power needs to come down 85% in cost to be competitive with renewables when looking into total system costs for a fully decarbonized grid, due to both options requiring flexibility to meet the grid load.
> Focusing on the case of Denmark, this article investigates a future fully sector-coupled energy system in a carbon-neutral society and compares the operation and costs of renewables and nuclear-based energy systems.
> The study finds that investments in flexibility in the electricity supply are needed in both systems due to the constant production pattern of nuclear and the variability of renewable energy sources.
> However, the scenario with high nuclear implementation is 1.2 billion EUR more expensive annually compared to a scenario only based on renewables, with all systems completely balancing supply and demand across all energy sectors in every hour.
> For nuclear power to be cost competitive with renewables an investment cost of 1.55 MEUR/MW must be achieved, which is substantially below any cost projection for nuclear power.
Or the same for Australia if you went a more sunny locale finding that renewables ends up with a grid costing less than half of "best case nth of a kind nuclear power":
Pretty much all of the dozens (56!) of reactors France built in the 70s took 5-6 years between construction and operation. How much weight do you give those dozens of builds, vs a couple one-offs? Yes, the fact that we've waited 40 years to build another single reactor is part of the problem. But it gets faster if you don't do crazy one-off designs and you get some experience building more than, you know, one.
If your hypothesis was true (cost overruns and delays are caused by lack of recent build experience in every western nation, and would immediately vanish if we started building more reactors) then we would expect to see the same clear trend of decreasing build time with experience in the past.
But in fact early reactors (e.g. Marcoule, or Tricastin) got built faster than the later ones (=> possibly because Chernobyl happened, and everyone reevaluated risk). So that hypothesis is clearly wrong.
And even if build times of under a decade were a realistic target for US/Europe (which I see no indication for)-- how many plants would we need to build first to hit those improvements, in your opinion? If we needed just a single full planning/construction iteration, then we could expect those miraculous 5-year-on-budget-reactors to come online in like 2050. Thats basically the best case scenario (!!). Might as well wait for fusion reactors then (that was sarcasm; fusion reactors are gonna be irrelevant in fighting climate change for exactly the same reasons; too much complexity, too expensive, too slow).
Vogtle actual costs are already competitive with actual renewables costs, and build times are vastly better than the 25 year timeline you're calling a "best case scenario" here.
> Vogtle actual costs are already competitive with actual renewables costs
No. At $160/MWh? Maybe competitive with renewable costs 15 years ago. Most certainly not with renewable cost now (and those still trend down).
> build times are vastly better than the 25 year timeline you're calling a "best case scenario" here
That timeline is for finishing one plant at currently speed, and then planning and finishing a ton of new reactors within 5 year buildtimes directly after. But I see zero confidence that this is actually realistic, not even from the the companies that just finished building Vogtle...
Not quite, dear customer: You own the motionless hunk of plastic, and you only rent the functionality. See also streaming hardware, BMW heated seats, textbooks, and home security systems.
Individual change has been promoted for 40 years if not more. Promoting it isn't fixing the underlying problem, though I'm sure it makes some people feel better about where they stand on the self-sacrifice ladder when they have others to look down on.
It has been suggested that the whole individual change movement was created to distract the populace from the actual big polluters. Like answering the question of "Why do we keep extracting coal and oil?" with "Why don't you look at your own consumption and reduce that first".
Salmon farms need to go. They are a plague to the wild salmon populations.
Here in BC Canada, our local area wild salmon populations are finally recovering after the closure of the local salmon farms.
They crowd fish in biologically hazardous conditions, then pump them with antibiotics and chemicals to keep them alive. All this in open water pens shared with the rest of the ecosystem.
One better solution seem to be inland salmon farms.
Wild salmon are too critical to many food chains both in the water and on land.
However, no mention of this basic head scratcher in the docs. :rolling_eyes: