I know this guy has a Nobel in physics and I don't, but I've spent a really long time in the solar business, mostly in R&D but also on the financial side, and even have some direct experience with concentrators. This article is mostly puff, but still something about the guy's claims seem off to me.

It would be easy to say, "yeah, the guy's got a Nobel, but he's 96 years old," and write it off to some sort of senility or dementia. The article does not give me the feeling that's what's happening.

It would also be easy to say that the idea of concentrating light for solar energy has been around for a long time and assume that he can't possibly have an original idea on the topic. I can't say whether he really has an original idea, but I can say that if it simply collects direct sunlight from a large area and focuses it onto a small area, it doesn't matter if the reflectors and/or lenses can be made for free, the economics probably will not pan out (and at the very least, it will not be the clear-cut universal solution to cheap energy that he seems to think he has). If it does something new -- collect both direct and diffuse light and focus it on a small area, or shift wavelengths so the light that reaches the solar cell can be used more efficiently -- then maybe he's on to something. Simple, non-concentrating solar using flat-plate panels is already so cheap, though, that I'm not sure that even that will bring a clear-cut economic case, especially if the reflectors can't handle things like heavy wind or snow.

Which brings me to my final thought: I suspect this guy is working with an outdated version of solar economics in his head and simply doesn't understand the current market. He clearly knows his stuff as a physicist, so unless he is senile this seems like the most likely scenario to me. I won't rule out that he has some revolutionary idea, but even if he does, the economic impact it can have is limited simply because the costs he can cut are already low. There's a certain way in which this reminds me of article I read where somebody invents a near-perfect antireflection coating and thinks they're going to revolutionize solar cells, which already have such a low reflectance that there is little to be gained there, especially if the new technology is not essentially free.

Not only that, it's not unusual for an elevator or escalator to be out of order in Parisian metro stations (or shopping malls, or pretty much anywhere else you might find an elevator or escalator). I am intimately familiar with the Paris public transit system for the able-bodied, but have also had to navigate it with somebody who is wheelchair-bound. We made a conscious decision to avoid the metro and use buses exclusively. They work pretty damn well for the physically disabled, and all of the discounts mentioned here will apply. There can be serious delays at rush hour, but really the biggest delays came when she and her family decided to use Uber -- the buses (and licensed taxis, for that matter) can use express lanes that, while still congested, are faster than anything but a train during peak traffic.

There is quite a big modernization of the Paris metro underway. It is focused largely on expanding the network, but also on updating existing lines. I'm really hoping they will improve accessibility of existing stations (and maintenance of elevators and escalators) in the process.

I don't know who you were working for, but I've participated in the construction of several solar factories in the past 20 years and none of them ever took as long or cost as much as what you're describing. The last one I worked on was a cell fab that took 8 months and cost less than $150 million for 1200 MWp. Its companion module fab cost around $50 million. And that's for a name brand with a solid warranty and reputation for high quality. I'm guessing the plant you worked on was more than a few years ago -- these days, 500 MWp is considered small, and even 20 years ago 3-1/2 years would have been a ridiculously long construction schedule (unless you're talking about polysilicon, in which case it would have been long until recently, but not ridiculously so).

Now I know you've been out of the industry for awhile -- the last time average solar panel prices were as high as $0.80/Wp was around 2012-2013. Even premium manufacturers like SunPower and Panasonic can't get $0.80/Wp right now. Average prices at the moment are below $0.30/Wp and, while manufacturers are struggling to be profitable at those prices, most of them are generating positive cash flows. End-user prices are a bit higher in the US because of tariffs on Chinese solar cells and modules, but nobody is spending anywhere near $1.50/Wp to make a solar panel these days.

The vast majority of thin-film solar panels are cadmium telluride solar panels made by First Solar. Yes, they use cadmium and tellurium, but the safety hazards are mostly occupational -- CdTe is a compound with properties different from its chemical elements, just like water is made from hydrogen and oxygen yet is not explosive like its components are (it's even used to put out fires). Indium and gallium are used in CIGS solar panels, but those make up a tiny minority of thin-film solar panels.

Oh, and thin-film solar panels make up less than 5% of the total solar market. More than 95% is silicon-based.

The nastiest things in that "brew of chemicals" are silane, phosphorus oxychloride, some of the solvents that go into the silver paste used to screen print the metal contacts on the cells themselves, and perhaps some of the chemicals that go into making the vinyl-based encapsulants and teflon-based backsheets in the final solar module. People like to draw parallels between solar cells and some of the nasties used to make microchips -- no idea if that's what you're doing -- but it simply doesn't work that way.

That's not to say the industry is perfect -- Jinko Solar, for example, faced violent protests 8-10 years ago after it dumped waste into a river and killed fish that the locals depended on for their livelihoods -- but the waste products are generally far less potent than the microchip industry. They're also far higher in volume, and it's an industry that depends on having a "green" image, so the risks are often overstated.

Oh no, I'm with you. One of the better cocktail bars in my city is a short walk from my apartment, but I've been there exactly once while I regularly travel halfway across the city to other bars, many of them not as good. And the simple reason is that I can't have a conversation and spend quality time with my friends because the place is so damn loud (and dark, for that matter, to the point that if you don't have a smartphone with a flashlight function then you pretty much need to have the waiter explain the menu to you, and, well, you can't hear him, so...). And this isn't some sports bar with shooters and cheap mixers, we're talking €12-15 craft cocktails. The main thing I remember about the place is thinking, "it's a shame I'm drinking this so fast, but I just can't wait to get out of here."

This reminds me of a story an acquaintance of mine once told me. She has a Ph.D. in statistics and has put in a couple of decades with a major biotech firm. A friend of hers was doing a Ph.D. in engineering and had some data to analyze, but he wanted to make sure his conclusions were statistically sound. He asked her to check his work and let him know if he had made any big errors. I'm sure nobody will be shocked to learn that he made some basic errors that non-statisticians make all the time (I think it had something to do with multiple comparisons). Once the analysis was done properly, none of his data showed the level of statistical significance he was chasing.

But did that stop him? No way! He went back and collected more data -- oh wait, no he didn't. He threw out the correct analysis and kept his original (incorrect) work for publication. And my acquaintance no longer does free statistical analysis as a favor to anybody because she thinks she'd be wasting her time.

You're still correct on the basic principles, but the figures you give for crystalline silicon cells describe low-quality cells these days. Typical warranties these days are 2-3% degradation in the first year and 0.5%/year thereafter. The vast majority of field data is not public and that that is public suggests that plenty of modules conform to this sort of warranty and plenty don't, but for what it's worth, reinsurance companies (who actually have access to the highest volumes of field data) are willing to take on the risk of underwriting such warranties.

Degradation in amorphous silicon has to do with the structure of the material itself, not the manufacturing process. That said, perovskite does have its own very serious degradation problem -- much worse than amorphous silicon, in fact -- that needs to be solved (or at least improved) before it will become practical as a major energy source. Even if it can be made cheaper than crystalline silicon, it's hard to see it gaining much traction outside of consumer gadgets and specialized short-term applications if the cells die after 5 years. Someone could still make money that way, but it wouldn't address much of our energy demand.

That said, what I've seen of this work is promising because the scientists are at least making an effort to address many of the issues that have to be solved to commercialize perovskite cells. There is increasing attention paid to that, but it seems to me that many researchers still prefer to chase headline-grabbing efficiencies. I wish them luck, but I do firmly believe they will struggle to create a successful commercial product if they don't make some strides on longevity.

We really shouldn't need adblockers, but we do, and this is precisely why. I don't mind a web site making a little cash from putting ads in front of my eyeballs -- it's exactly what publishers have always done with newspapers, magazines, and TV shows -- but when they waste my time or render a page unreadable then I'm done. Pages with delayed loading of ads or videos such that the text on the page is constantly moving after I have already loaded the page -- who ever thought that was a good idea? I understand why publishers scream about adblockers, but really, with many (and increasingly more) sites, if the publisher isn't going to build a reasonable web page then I'm either going to use an adblocker or I'm going to go elsewhere. Either way, the publisher won't make any ad dollars off me. I don't think I'm being unreasonable, but many publishers certainly seem to think I am.

Unfortunately, at work I don't really have much control over these things because my computer is locked down so we can only use software that has been vetted for security and privacy. Ironically, that means using Chrome with settings locked by the administrator to share maximum information with Google, no ability to install extensions like Privacy Badger and Adblock, and no ability even to install tools that make it easier to work with our approved office suite and project management tools. I don't think they realize just how much money they're paying me to fight with software that is supposed to be making me more efficient, but that's getting off-topic....

Private funding didn't come in for any of those until the government had provided so much support

Been there, argued that. Step one, where are your citations?

Railroads:

• Doukas, Kimon A. The French Railroads and the State. Columbia University Press, 1945.
• Dunham, Arthur. “How the First French Railways Were Planned.” Journal of Economic History. Vol. 1, No. 1, 1941.
• Pirenne, Henri. Histoire de Belgique. VII: De la Révolution de 1830 à la Guerre de 1914 (2nd ed.), Maurice Lamertin, 1944.
• Bain, David Haward. Empire Express; Building the first Transcontinental Railroad. Viking Penguin, 1999.
• Railway Guarantee Act of 1849 (Canada)
• Pacific Railroad Acts of 1862, 1864, and 1867 (United States)

And that's just the early industrial history of railroads. It was kind of a mixed bag of private and public investment. More recently, high-speed rail has been deployed almost exclusively by government-owned rail companies like JNR in Japan, SNCF in France, DB in Germany, British Rail in the UK, Amtrak in the US, Renfe in Spain, Korail in South Korea, and perhaps most importantly these days, China Railways, which was literally a government ministry until 2013. Even Thalys, which is nominally a private high-speed rail company, is 100% owned by SNCF, SNCB (government-owned Belgian rail company), and DB. For more on that I refer you to Wikipedia and decades worth of annual reports by the individual companies, I'm not going to link to them all for you.

Telegraph:

• Stover, John F. History of the Baltimore and Ohio Railroad, 1987 (yeah, the title suggests it's about a railroad, and it is, but it also includes the story of the first telegraph line in the US, which was funded by the US Congress).
• http://history.house.gov/Exhibitions-and-Publications/Electronic-Technology/Telegraph/
• Janson, Michael A. and Yoo, Christopher S., "The Wires Go to War: The U.S. Experiment with Government Ownership of theTelephone System During World War I" (2013). Faculty Scholarship. Paper 467.http://scholarship.law.upenn.edu/faculty_scholarship/467
• https://en.wikipedia.org/wiki/All_Red_Line (yeah, Wikipedia, but I can't find the book I'd rather refer you to)

Telephone: Don't forget that in most places outside of the US, the entire telephone system (including the telephone itself) was owned by the government until the 1980s.

• Janson, Michael A. and Yoo, Christopher S., "The Wires Go to War: The U.S. Experiment with Government Ownership of the Telephone System During World War I" (2013). Faculty Scholarship. Paper 467.http://scholarship.law.upenn.edu/faculty_scholarship/467 (yes, a repeat from the telegraph section)

What do you mean, "as it did in due time"? Did you somehow visit another reality and see if the internet would develop in the absence of government funding?

Now, that is rich. As if you have somehow visited another reality, where the Internet did not develop without it...

Holy missing the point, Batman! You might want to re-read your own comment. If you still can't understand my response, try applying basic grammar and logic.

And thanks for reminding me why I stopped bothering to try to engage in intelligent conversation on the internet.

That's nonsense. Private companies did fund and successfully built networks of railroads, telegraph, and telephone.

I think you might need to review your history a bit. Private funding didn't come in for any of those until the government had provided so much support that it reduced risk to "acceptable" levels. Yes, private funding made all of those networks pervasive, but government funding made them possible.

They would've built the current Internet, when the technology developed — as it did in due time.

What do you mean, "as it did in due time"? Did you somehow visit another reality and see if the internet would develop in the absence of government funding? Considering that literally none of the hardware that runs the internet developed without early government support, I don't see how you can know that.

With the possible exception for militarily-applicable research, no science should be government-sponsored. At all.

Well, we have government-sponsored research to thank for your being able to share that comment with us. Without government-funded science for both peaceful and military purposes you wouldn't have computer to type your comment on, nor an internet or World Wide Web to transmit it over. You not only wouldn't have a smart phone, you wouldn't have a cell phone, or any phone at all for that matter. Or even electricity, most likely.

You can't rely on wealthy investors and venture capitalists to fund science for which there is not a clear application, customer, or business model, especially if that business model does not lead to profitability or an IPO in a relatively short period of time. Thirty years ago the first web browser was still two years away. The first web browser that anybody has heard of was still five years away. The only networking business case for the rabble that anybody really imagined was dial-up service à la Prodigy, Compuserv, and America Online -- and those services largely kept customers inside their walled gardens and made it difficult or impossible to access the internet itself. Even after Mosaic appeared in 1993 (a government-funded effort, by the way) and people started to get their first taste of the web as we know it, it was still years before private investment grew significantly because people needed to get online for any of it to matter, and doing that required both public investment and new business models.

The usual suspects were first on the scene, of course: The first time I encountered a camgirl with a live video stream was in 1996....

When I left the US a decade ago, I was paying about $100 a month to Comcast for internet service and basic cable (no premium channels, no HD, not that it mattered since I didn't have an HDTV). Today in France, I pay €39.99 a month for basic cable, internet service, VOIP including free international phone calls, and all the mobile phone calls and text messages I can make (within France, international texts typically run me about €0.20 a month). My parents in the US, for that same bundle of services, are paying more than $200 a month.

I can't really see myself doing anything but streaming TV if I ever move back to the US. I enjoy TV, but I don't need it. Why should I spend a bunch of money supporting a business model that doesn't really serve me?

Indeed, because they add a lot of services into that "credit" side of it like for example travel insurance and kickbacks where customers appear to be "saving" money. In Europe there's actually a lot of alternative debit card systems run by the banks, like here in Norway there's "BankAxept". In Germany they have "Girocard". Without all the bells and whistles here in Norway at least they pay roughly $0.02/transaction, which is way below the cost of actually handling cash. So honestly, the only reason businesses accept cash here is because they must.

Perhaps that's true in Norway, but I don't think that generalizes to all of Europe. It certainly has not been my experience in France or (especially) Belgium. In Belgium I always make sure I'm carrying double the cash I think I need and I stop at the ATM every chance I get, because they don't have them on every corner like in some parts of the world. Even many of the businesses that are not cash-only will only accept cards that work on the local Bancontact network, which means that if you don't have cash you have to have a Belgian bank account, and some charge an extra percentage or require a minimum purchase (10€ is common) if you want to pay by card. Up until a few years ago you couldn't even buy a train ticket from an automated kiosk without cash or a Belgian bank card! That's been changing a bit over the last five years, especially in the bigger cities, but I don't know anywhere in Belgium that prefers plastic over cash.

Big cities in France are not quite so extreme as that, largely because they have so many foreign tourists, I think, but the countryside can be much the same as Belgium. Even in central Paris, though, I always ask what their minimum is before I try to pay by card. Even the big grocery chains have one, though it is usually only 1€.

Then contrast that with China, where at least in urban areas they've gone from a largely cash-based system to mobile-phone-based payments in a very short time, practically skipping plastic altogether....

Tesla does its manufacturing in Buffalo, New York, though the first panels just started rolling off the production line and it is not yet clear how good or viable a product they are. SunPower is based in San Jose, California, but its manufacturing is done in the Philippines, Malaysia, Mexico, France, and (soon) China (though Chinese production will be for a lower-efficiency product than its flagship). Tesla is not affected by the new tariff, though it did oppose it, but SunPower is.

China's tariff on American polysilicon was imposed in retaliation for American tariffs on Chinese solar cells and modules. Prior to 2013, the US actually had a trade surplus of quite a few billion dollars with China on the basis of the massive amount of polysilicon shipped to Chinese customers; since the US imposed tariffs on Chinese solar products and China retaliated, China has gotten the upper hand on the trade deficit (and found the motivation to learn how to make pretty good polysilicon pretty cheap).

Any evidence for Daqo getting free electricity? I'm familiar with them in a way that few people are and I've seen no evidence of such.