(For the record, I didn't plan to go reading your Wikipedia page, I just searched your name and "fusion" so I could go read the papers you had written, and this is what came up instead, lol)

Fission is perfectly good in space. What, you worry about the radiation?!

Oh for fuck's sake, you cannot be this ignorant.

Fusion in space (beyond the power side, where it's in general higher temperature (higher Carnot efficiency / easier to radiate) and lower mass than fission) is about being able to exhaust fusion plasma as a high-ISP rocket engine.

ORLY?

https://en.wikipedia.org/w/index.php?title=User:Maury_Markowitz&action=history

Because your wikipedia user page started out saying that you're a programmer working at a hedge fund who got into programming by working in tech support.

I work at a medium-sized (for Canada) hedge fund during the day, primarily writing the program they use to enter and track orders. I'm formerly a Mac guy, but holding a day job pretty much means you've got to work on the PC, and so I do. I can't say I really mind it though, and I have to admit that Microsoft Access does the job well.

I got into programming in a roundabout way, originally working in tech support for FirstClass. That was one odd company; there were the programmers, and then "everyone else", the peaons. I never managed to break into programming there, the barrier to entry was just too high. Then back in '97 Apple Computer announced it was buying NeXT and using OpenStep as the next Mac OS X. So I started looking into OS and got completely hooked, posting about it a lot on the UseNet. Then one day I got a email from a developer in Toronto who wanted to hire me to help him write a program on OpenStep, but I declined, saying I liked my job and didn't really have that much experience anyway. The next day I got laid off. The day after that I worked for him. The rest, so it goes, is history.

Then when you started editing fusion-related pages, you changed it to say:

The quick-n-dirty description of me is "failed physicist" - I took physics in U but I completely bit at the heavy math. So now I'm a programmer, like all the other physicists out there. Eh, I don't mind that too much. I'm also a pilot, so unsurprisingly most of my edits are on science, tech or aircraft.

Then you later edited it to say:

The quick-n-dirty description of me is "coding physicist" - I took physics in U, and like so many others of my era, today I'm a programmer

Now here you're:

A physicist who has been writing about fusion since my 3rd year E&M thesis

Go home, poser. You're a programmer who took some physics courses in school, failed them, and are now pretending to be a subject matter expert.

The cost of a fission plant outside the nuclear island - that is all the things like steam generators, turbines, cooling loops, etc. - is about 60% of the total cost.

It's even more than that on average. But this isn't a fission plant. It's much more akin to a coal or NG plant than a fission plant. ARC is dealing with superheated steam (540C, like a coal plant), not the ~300C or so you might get in a fission plant (fission plants require enormous turbines per unit power, and MSRs). Plus you also have to reject a lot more heat due to the lower thermal efficiency (*on top of* the much greater steam volume). You're also not having to meet nuclear quality assurance standards on the site - backup generators, emergency cooling, and a whole slew of other things, which are not only fundamentally expensive, but you often can't use off-the shelf systems.

Bringing up the cost of this stuff from fission plants is nonsensical. This has nothing to do with fission. If something goes wrong, the reaction stops instantaneously, and the only thing you do is damage your inner core, which is a consumable item anyway. It just means moving up your maintenance cycle. Your balance-of-plant costs are coal-like.

Assuming MIT's ridiculously low estimates of reactor cost

It's only "nonsensically low" because you don't like it. There is nothing unreasonable about it relative to the size of the undertaking.

This isn't ITER where they're employing a veritable army of scientists and engineers on government contracts for decades as a jobs programme.

PV systems in the US currently cost about $1/W

[Capacity Factors have entered the chat]

"1W" of solar nameplate capacity averages 0,24W in the US.

A fusion plant, when mature, can be expected to have a capacity factor similar to a fission plant, e.g. downtimes mainly just once every 1-2 years for maintenance (in a fusion plant, replacement of the inner core structure). 90%-ish. Otherwise, it's just constant (pulsed, accounted for in the 400MWe) generation.

Also, for the record, 1W-ac of nameplate solar power in the US averages ~$1,60,W-ac not $1 (as of 2024 at least). Don't compare W-dc with W-ac (also, even $1/W-dc, while "in the range" in the US, would be a good price - 2024 average was $1,22/W-dc)

Also, I don't know how to break it to you, but not everywhere on Earth is the US desert southwest. Hey, I live in Iceland - want to take a wild guess how well solar is taking off here? Even in the summer fixed PV sucks because the sun does an azimuthal 360 around you (and our peak electricity demands are in the winter). Also, PV isn't compact. You're not going to power a large ship with PV. You could with a fusion reactor. And we're not even bringing up space here.

The development cost is, compared to the amount that gets invested in the grid every year, basically in the noise threshold. It's well worth it.

With storage, that goes to about $2/W.

By "storage" you don't mean "able to handle a dunkelflaute".

Don't get me wrong, I like solar. But this is a terrible argument against fusion.

Oh, lol, I just noticed that the person you cited was BUSSARD.

Yes, I know most people here know him as the Bussard Ramjet guy. But he was also the Polywell guy ;) He's was hardly the guy you'd want to be citing for "mainstream" fusion commentary even back in 2006.

(Also for the record, Bussard Ramjets don't work either).

more like the type most understood to not work.

Simply false. The Q factor is eminently predictable with scale. It is by far the most predictable form of high-Q factor fusion (outside of gravitational, and we're not going there any time soon ;) )

as Robert Bussard said in his famous Google TechTalk in Nov 2006

What, you mean BEFORE we got commercial-scale HTS magnets that scale down the size requirements by an order of magnitude?

Also, pointing to things like ITER to say that cost-effective fusion is impossible is like pointing to the ISS and saying SpaceX is impossible.

And also pointing to a single person's two decades-old view as if it represents a whole field, today (FYI, it doesn't, at all) is pretty damned funny.

Candiakenvirchusevania?

All these commenters who think they're so smart coming out with the same "Fusion power is 20 years away and always will be, har har har!"-quip who don't know a damned thing about the field and its progression is so tiring. One error with neutron measurements at ZETA before we even knew what we were doing, and the entire field was turned into a permanent joke, even as Q factors continued to climb almost monotonically. The press had their story and now we're cursed with an endless stream of these people.

Yeah, so, this is not true.

First off, turning it "to powder" is hyperbole; metals just become increasingly brittle.

Secondly, claiming that there's "no solution" is not just wrong (there are many), the particular solutions used by Commonwealth are literally discussed in the papers that this Slashdot article is about. Specifically, they use a molten FLiBe breeder blanket to absorb the fast neurons, which also breeds tritium. Since it's molten, there are no "structural" issues with it at all. The inner core (mainly tungsten) does need periodic replacements (every 1-2 years), but the reactor is designed to be easy to open up for swap-outs. It is treated as an expendable consumable, and is melted down and recast/rebuilt for the next replacement. In terms of complexity, cost, and downtime, it's probably roughly on par with fission reactor maintenance periods, perhaps superior.

Third, there are many types of magnetic confinement fusion, not just magnetized target fusion. These are less mature than tokamaks, and generally considered more longshots. Even ignoring that the fusion itself is more challenging, they trade something relatively simple - materials science and swapping - for something much harder (immense mechanical and fluid dynamics challenges)

Fourth, if you really hate neutrons, there are also aneutronic fusion designs. Again, though, less mature.

Assuming sparc (no power) costs $1 billion, then guessing that arc costs $5 billion and makes 400MW. You could install about 2GW of sea based wind for the same. With such a huge power surplus over fusion you could probably melt rocks to store power for the still days

1) ~$5B is about right for the first ARC plant, but that's to be expected, because first-of-a-kind plants are always much more expensive. Nth-of-a-kind for ARC is expected to be about $2B.

2) Wind is variable load, not baseload, not load following and certainly not peaking. Its power is worth much less.

3) If you want your wind farm to be able to get through a mere 5 day dunkelflaute and guarantee a steady 400MW output, then, with a 40% round trip efficiency, you have to store 120GWh of thermal energy. Even if your storage is a mere $25/kWh, which is extremely optimistic, that's $3B. And since your wind farm is throwing a lot of its energy away to the losses inherent with thermal storage, you're looking at $5B for the wind farm. And then there's $500M for the power block on top of that. You're looking at a $8,5B project.

(Of course, thankfully, that's not actually how we build out high-renewables grids)

HIGH temperature, not ROOM temperature.

*Facepalm*

God, Slashdot comments sections are so embarrassing these days.

For YHVH's sake, first off "suggest" is not Commonwealth's wording, they wrote five bloody peer-reviewed papers. You're criticizing them based on a word that a Slashdot author chose, likely without even thinking about their wording.

Secondly, there's nothing mystical about tokamak fusion, it's the most well understood type of fusion out there. The scaling factors are well understood. What the "entities" whose "corpses" litter the field didn't have was high-temperature superconducting magnets, as commercial-scale availability of HTS tapes only emerged in relatively recent times. These let you double the field strength. Under tokamak scaling factors, doubling the field strength lets you get the same Q factor at around 1/10th the volume.

There's many other interesting aspects of note, but at a fundamental level, that's all you need to know.

So your idea is that Russia takes Estonia, and then just decides, "Meh, I'm good, that's enough"?

You understand that Estonia is part of Europe, right?

You understand that Estonia is part of the EU, right?

You understand that Estonia is part of the NATO, right?

Also, FWIW, the Russian population of Poland is the size of a large town or a very small city

Russia is more than happy to use "pro-Russian slavs" even if not ethnic Russians. And FYI, 20% of Estonia's population is ethnic Russians.

There actually are ways to poll people in repressive regimes. A classic one goes like this: instead of saying "Do you support President Putin?", you give them a list of a bunch of world leaders - say, Putin, Trump, Modi, Erdogan, and Macron - and ask, "How many of these leaders do you support?" And then you give the next person a different set of leaders - say, Putin, Xi, Starmer, Khamenei, and Merz - and again ask, "How many of these leaders do you support?" You can then statistically disentangle the level of support for Putin, without any individual having to ever say whether they support Putin or not.

Given what Ukraine is managing to get past Russian air defences, do you think Putin is going to nuke Europe and risk French nukes hitting Moscow and Petrograd?

That's not how this works; Russia comes at more powerful opponents sideways, with incrementalism, division, and hybrid warfare. They try to get pro-Russian candidates elected so that they can veto collective action. They have "accidents", such as missiles flying into your airspace, to probe how you'll react, and if the reaction is insufficient, probing ops can go further, or become normalized so that you get used to it and then don't freak out when they go a bit further the next time. They famously support "pro-Russian rebels" (quotation marks definitely demanded) in countries to try to break off chunks, and encourage secession movements. Russia prefers to work piecemeal, tearing apart countries by chunks, and only launching large-scale invasions when they think (rightly or wrongly) that they can get away with it.

Do you honestly think that France would fire nukes at Moscow - and thus face nukes raining down on Paris - if quote-unquote "Russian Separatists" with suspiciously good arms and military training took a chunk of one of the Baltics or Poland? Of course they won't, and Russia knows that. And then, "Oh, we're just sending aid to the oppressed people in the separatist regions!" "Oh, the separatist regions have requested help from us, we're sending advisors!" "Oh no, the separatist regions were "attacked" by their oppressive government, we must launch a counterattack in response!"

This is hybrid warfare. Nobody fires nukes in response to hybrid warfare. Meanwhile, Russia gets incrementally more powerful with each region it captures. A couple decades ago Chechens were bloodying the Russian army. Now they're dying on behalf of the Russian army in Ukraine. Everywhere they take represents human, industrial, and mineral resources. People are constantly propagandized to and kept in a state of poverty from which only service to the Russian state can rescue them.