tsjackson

I work in the industry. While this is theoretically true, in practice it is not.

First, bonds for decommissioning solar sites are required by many programs now. These usually cost 3-5% of total initial investment. Not insurmountable by any means.

In reality, these bonds are unlikely to ever be utilized. Most sites will never be decommissioned unless the owner wants to redevelop the land for something more profitable. The interconnection rights alone are worth way more than the cost of decommissioning.

Recently, repowering solar sites has become a hugely profitable endeavor. Most projects have a 20-30 year power purchase agreement, often at a fixed price per MWh. That initial price was set very high based on the massively more expensive cost of deployment (equipment and install cost about twice as much 10 years ago). Squeezing 20% more capacity out of these projects by repowering these sites with more efficient modules and new inverters, is often a windfall, given the falling costs involved.

That's not to say there are zero issues like the ones described. Like most early stage technology-heavy industries, early solar projects often utilized a lot of "innovative" technological strategies that dead-ended and are therefore much harder to maintain.

But overall this is just not much of an issue in practice.

10k is a reasonable guess for how much tax liability is needed, and it's not a huge issue, but it is an issue. Retirees and lower-middle income folks often pay less than that each year, as do folks with even lower incomes who inherited their homes.

Particularly if you go the hardware route, might be worth exploring this as an option for utility scale solar farm inspection. Quadcopter drones often have insufficient range for the larger farms, and solar companies are increasingly making full site scans a regular part of the O&M process (they have has been used in commissioning larger sites for years).

I actually 100% agree with this comment, except the "only being done to maintain appearances" part. A blunt instrument is better than no instrument. By raising interest rates, the fed is reducing average inflation by cooling the market across the board (though unfortunately with very little direct effect on the primary driver of energy consumption, which is very inelastic). Without cooling the economy a bit, normal inflation plus the supply side drivers could lead to hyperinflation, and/or stagflation.

Supply chain issues won't sort themselves out for 3-4 years, possibly more - it can take at least that long to get a new domestic semiconductor chip fab or solar panel factory from the idea stage to full capacity. And if you are a business, the level of uncertainty as to what 4 years from now will look like makes a huge investment like that less than desirable. (Source: I work for businesses in these spaces).

Businesses just aren't as nimble as we were led to believe, and it's going to be a bumpy few decades in all likelihood, assuming China stays on the path of no-dissent nationalism and the U.S. stays on the path of re-industrialization.

In the long run, we need to transition the energy grid to electric/renewables/storage as fast as possible to get off of the fossil fuel roller coaster that has caused every major inflationary event. In the medium-term, we need to reduce impediments to building physical things in our country, so that businesses can respond more quickly to increases in prices by increasing supply.

Let's take an extreme example. If banks offered everyone 0% interest mortgages with 100 year terms and no credit checks, how many more people would want to buy a house? A lot! It would take decades for builders to build enough houses to meet that demand. In the meantime, house prices would go up by a factor of 10, along with wood and other building materials. Meanwhile everyone who already owns a house is cash out refinancing their now multi-million dollar home, and suddenly everyone you know who was a homeowner at the beginning of 2023 is now a million dollars richer and spending it quickly.

The same with businesses. Let's say you're a growing business. Things are good and you've been investing your 10% profits each year into hiring. Now banks decide that all decently profitable businesses can have that same 0% interest mortgage with 100 year term. Why not double, triple, quadruple your team? You could achieve your goals so much faster! But then the banks are offering all of your competitors the same deal. But there's not enough talent to go around. Suddenly you're in a bidding war for decent sales guys and the starting price is a million dollar salary. And those million dollar sales guys are spending their salary, competing with other million dollar sales guys for shit they don't need - the price of everything goes up.

This is an extreme example to illustrate WHAT JUST HAPPENED with record low interest rates. The economy was being heated up by very very cheap money. Inflation started getting out of control.

Now, if you were in the above hypothetical scenario, you might say "Hey maybe we shouldn't give out all those crazy loans, people are going crazy with all of this money, and it's kind of fucking everything up." And you would be right.

And so is the Federal Reserve.

Also, reducing spending is even slower than reducing taxes.

The vast majority of government spending is on Medicare, Social Security, Medicaid, and defense spending. Politically those are untouchable, mostly for good reasons - reducing any of them will result in people literally dying.

The vast majority of what's left is hugely impactful high ROI activities like scientific research, infrastructure projects, and other basic good governance activities.

Too slow. While I fully agree that taxes have gotten out of whack and should be raised, raising taxes would have an impact a year from now (way too slow of timeframe to manage inflation). Additionally, practically speaking, the government would have one shot and no realistic way to quickly correct if they raised taxes too much or too little. Finally, there's so much uncertainty in the market when major political decisions like that are afoot, that you risk doing harm just from the perception.

The fed has the power to act quickly. They raised rates a little on almost a monthly basis last year. Each was a little experiment. If they raised them too much, they could reduce them the next month. If they raised them too little to fully counter inflation, they could continue raising them.

Finally, the issue isn't getting "money" out of the system - it's getting purchasing power out of the system - reducing demand. And most people are buying 5-20% of houses, banks are buying the rest. Most large businesses aren't paying cash reserves to pay employees, they're using debt to pay those salaries.

Lots of smoke in the comments here. The Fed is doing what needs to create relative stability. Uncomfortable, but real. Demand is outstripping supply and prices are going up. The least painful option is raising interest rates. Alternatives are hyperinflation, (very bad), or various price fixing schemes (which have literally never worked despite many attempts and are even worse in the ultimate outcomes). There are lots of reasons why this is happening, and none of them are related to a "fake economy:"

People and businesses came out of lockdown with saved money and basically free loans burning holes in their pocket which caused a spike in demand (Least important, probably no longer an issue)

Businesses came out of lockdown with a diminished staff and a ton of new uncertainty (much more important, takes a while to recover for some businesses that are planning production multiple YEARS in advance).

Deglobalization/ U.S. national re-industrialization, started by Trump, continued with Biden, which will increase prices on pretty much everything. This is both a reasonable response to the issue, and makes the issue worse in the short term.

There's a hot war with a major energy producer, Russia, which will increase prices for every product where energy is an input (almost every product).

The biggest manufacturer in the world, China, has randomly been shutting down factories and whole metropolises for weeks at a time for the last several years. We just got a correction in this regard, but it will take time for the supply side of the equation to ramp back up, especially given all of the moving parts and uncertainty outlined above.

Bottom line - lean supply chains function well when everything is stable for a longish period of time and it looks like it will continue to be for a longish period of time. In unstable/uncertain environments, supply chains break down, and supply can't keep up with demand, and the government can't keep handing out free money without causing prices to hyperinflate.

I'm really curious about this.

Hypothesis 1) Tik Tok believes AI/machine learning algorithms themselves are converging across fields, and the company is finding that its internal AI innovations can be applied to biology and materials sciences algos and, possibly, vice versa. We'll call this one the "data are data" hypothesis.

Hypothesis 2) Tik Tok has decided that it's an AI company, not a content platform company. This is subtly different, as the locus of intersection between social media and materials sciences is in the company's human capital, not the algorithms used. Call this the "Data Scientists are Data Scientists" hypothesis.

Hypothesis 3) Tik Tok, as an avatar of the CCP, is deliberately adding fuel to the narrative that its a national security threat, thus baiting Biden to shut it down, which would be hugely unpopular with one of his critical voting blocks. Call this the "Troll Biden" hypothesis.

All of these hypotheses seem really important from a geopolitical perspective. If social media companies become our primary engines of applied sciences, that would be a pretty huge shift. If the CCP is using Tik Tok to use cultural popularity to play chicken in the trade war with the US government, that's a pretty major development as well.

Any other ideas? Am I missing something obvious?

Modules degrade in production really slowly (<1% per year) and are often used for 40 years. They're generally replaced quicly if they break due to manufacturing defects or impact damage (hail, tree branches etc.). A lot of the recycling of solar panels is for panels that are a decade or two old but in perfectly good condition. What's happening is that new modules have increased 30-40% in efficiency over the same time frame that the old modules have decreased by 10-15%. So its cost effective to upgrade them in certain cases - generally where the utility will allow it.

The other factor here is that many agreements between a solar power and a utility that buys the power only have 20 year terms. Generally, there's a strong incentive to renegotiate at the end of the term, but frankly, that renegotiation is kind of a mess in practice. It depends on the policies of the state, the utility's interests, the ISO market, etc. as to how that ends up working. Every solar farm is working in at least five overlapping regulatory environments - local, state, utility/retailer domain, the ISO or regional grid, and the federal regulatory environment. Decentralization is nice in theory, but definitely makes it difficult to scale the widespread change that's required right now.

Thanks for checking me on my U.S.-centrism and your comment as well! For the record, I think Australia's market is a bit younger even though it's already a bit larger as a percentage of power consumed than the U.S.'s thanks to some amazing solar resources in the desert. I'm fairly certain that all of this gets much easier for everyone at scale - it's just a matter of waiting it out until there are enough modules ready to recycle to generate regular revenue for recyclers, so it's likely just a waiting game.

You're right! Thanks for the correction. Thin film modules use Cadmium Telluride, which includes two rare metals* (not rare earth* metals). I want to note that there are some similar supply concerns with each of those rare metals. But in any case, my larger point is that thin film modules represent less than 10% of the market, and are in no way critical to solar's success.

I work in the solar industry, and very few large projects are turfing solar panels anymore. There are at least three companies that will usually bid against each other to purchase used/broken solar panels for recycling/reclaiming materials.

One of the early issues with recycling companies scaling is that solar modules don't break very often, so there hasn't been enough volume to get the industry off the ground. Solar modules are generally good for 10-40 years, so we're just starting to get the first generation of decommissioned plants (which by the way, are generally being repowered with more efficient modules).

Same with wind turbines. In any case, outside of the valuable heavy metals, landfills really aren't that huge of a problem, despite consumer focus. Decommissioned landfills are already a hot commodity among solar developers in the Northeast for instance because they're great, relatively flat, centrally located land that you can build a solar farm on. So as long as we're succeeding at reclaiming heavy metals, the waste generation component is pretty trivial. They're really just part of the cycle.

Finally, the decommissioning cost of solar plants is usually bonded in with a utility PPA to be borne by the project company, just like with nuclear, so it is indeed a fair comparison.

I agree regarding our regulatory environment for nuclear being counterproductive (it's counterproductive for wind and solar too, though to a lesser extent). However, even in positive regulatory environments such as France, Nuclear costs 3-5 times as much to build on a $/MWh basis and takes much longer to site, permit and construct. There may be a small role for base-loading nuclear in certain areas that have poor renewable resources, but it otherwise rarely makes sense, regulatory issues aside.

There's plenty of space - deserts of the southwest for solar plus offshore wind could easily suffice to produce all of the power needed in the US by themselves. The issue is lack of grid transmission infrastructure, which is a reliability issue during times of extreme weather independent of whether it's a renewables based grid or not. As for materials, there are some rare earth metals that are used in some solar panels, but many others don't require them. If market conditions dictated, use of those materials would shift on it's own.

There's a weird sociological phenomenon where people declare that "renewables won't get us there" without any real justification.

A recently announced deal in California would build out solar for $20/MWh compared to your $200/MWh nuclear worst case. Renewables can and will get us 90% of the way there. I recognize that a renewables based grid will require massive investments in grid infrastructure, which are long overdue anyway, as well as eventually storage/batteries, which is still on an exponential cost reduction curve. Bbut given the massive cost difference of generation, that would still be a dramatic net benefit.

Particularly when nuclear plants take a decade to build compared to a year for a solar plant.

I get that nuclear is cool, particularly fusion, but I really don't understand this "renewables won't get us there" argument beyond that.

The cost per kWh of solar and wind is a third of natural gas and nuclear. It's not even close. Filling the gaps left by intermittent production with storage is rapidly decreasing in costs, and will soon be low enough that Solar/Wind/Storage will outcompete Natural Gas or nuclear 95% of the time. There is probably a small role for other sources in filling capacity gaps during uncommon weather events (i.e. natural gas peaker plants during heat waves), but this blog post is a lot of complicating noise in what is a fairly simple economics of implementation problem.

Psilocybin (from mushrooms) and MDMA are both fast tracked for FDA approval as adjuncts to therapy. Psychedelics are poised to become the most common psychiatric treatment in the world, with ongoing trials showing huge effects for treating PTSD, addiction, and treatment-resistant depression. Despite a lot of therapeutic promise, LSD has not received as much research attention, primarily because of its duration of effect. Having therapists present for up to 24 hours (compared to 6 for psilocybin) is prohibitively expensive. A true off switch would facilitate its study and use in treatment.

I've been undergoing a similar transition in thought, likely triggered by the same events. I'm struggling to reconcile the competing lenses.

In our current political reality, anything that is not desirable to Republican elites will be argued against as if it were simply a technical "Mistake." No matter how many times they're out-argued, they continue arguing on a technical basis. "Tax cuts to the rich increase jobs and wealth for everyone" and "Climate change isn't being caused by human action" are probably the two most glaring examples of trash arguments that should sink under the weight of overwhelming evidence to the contrary, but they may never go away. And it seems that more and more political arguments are increasingly in this camp, emboldened by the persistence of these incorrect assertions.

Politicians/elites are blatantly arguing in bad faith at least some of the time. They won't be convinced by any argument or evidence, no matter how damning. They are getting away with this because outside of hard science, nothing can be "proven" and they will surface any doubt in an argument and simply magnify it. At some level, it's a social hardship that owes to epistemology and the limits of our knowing.

Of course there are still difficult technical questions of governance as well. Policy making is not easy. But at this point, it's extremely difficult to distinguish between arguing against "mistakes" and arguing against something that is simply inconvenient to the arguer. On the surface they generally look the same - like a person in a suit making a technical argument. The trash arguments are poisoning the well, making it impossible to share a space for discussing the truly difficult questions with those who disagree.

We live in an era of noise, and we are in desperate need of better filters. How do we detect an argument made in bad faith? How do we respond once we know an argument can't be won? I don't think we have good answers to either question.

BMAA is the neurotoxin (implicated in Alzheimers and Parkinsons). It also bioaccumulates over time, and potentially up trophic levels. So definitely not fish wonder food until there's some serious testing done.

Attention has quickly become our scarcest and most valuable resource, and filtering out noise is one of the most valuable services that can be provided. The no limits, "let the good in with the bad" approach is just not realistic at this point in human history.

Unfortunately, I don't think that sociologists are going to find the answer alone. It will take others who can engage the theory, but not become swept away in it. What's most troubling is that Sam Altman is just the sort of person who might be able to add value to the situation by taking a complex issue and thinking really hard about how to deal with it. We need to develop nuanced mechanisms of combatting misinformation and filtering out truly impoverished arguments while letting in unusual, perhaps even incorrect, but potentially valuable ones. Instead he merely repeated dogma of a "marketplace of ideas" model of the world that is clearly inconsistent with the evidence all around us.

This is, at least in part, a technical problem. Zuckerberg seems to be waking up to the role of the failed marketplace of ideas in creating the political monstrosity that we are dealing with now. He seems to be at least thinking through how to deal with it. But it's going to require a lot more than one organization to figure out how to communication must change in an era where the value of talk is so much lower and the value of the attentional resources required to truly listen are so much higher.

Techno-optimist drivel is at least as unsavory as the overly sensitive, anti-technological kneejerk reactions that this article is lamenting. It's absurd to write off real problems with technology because tech does good things as well.

I am having trouble following your jump from shutting down trolls to sacrificing our souls. If you think the loudest person should get a platform, I encourage you to spend some more time in (non-HN) comment sections.

There is nothing in my statement that suggests that every viewpoint that disagrees with my own should be censored. What I'm arguing is that the world has gotten a WHOLE lot more complex when it comes to mapping peoples' understanding of the world. The signal to noise ratio has turned upside down, and society as a whole has become more volatile as a result. We have to take the threats of this new reality very seriously and avoid broad generalizations.

I think that we can collectively come to a reasonable agreement about what should and should not pass as discourse vs. plain trollery. If someone is parroting "facts" about Barack Obama's secret Kenyan birth and closed door Muslim religion, all of which have been refuted, and cannot offer a standard of evidence that would be sufficient to change their mind, they aren't engaging in a debate, they are simply spreading a lie. That lie, and many others like it, have recently shown the ability to propagate propelled solely by regular repetition. Is censorship the only way to provide a pattern interrupt? I don't really know, but I do know that it is TERRIBLE for a democratic and free society to have huge portions of the population very certain about opinions that are totally, demonstrable false, not to mention that these same opinions also incline these people to be less tolerant, less trusting, and more aggressive. I don't think we should rule out some level of censorship, at least at the interpersonal level, for thoroughly refuted ideas. I'm not saying we should have government intervention, I'm just arguing that you should have every right to tell your angry, misinformed uncle to stop spreading lies at the dinner table.

On a personal level, my own standard is that you earn the right to have a discourse by being willing to let go of your viewpoint if it's refuted, in fact that's part of why you are entering into a conversation - to learn what you are wrong about. Strong beliefs, lightly held. The right to speak comes with a responsibility to engage honestly.

The main thrust of the argument seems to be that we should do our best to stop censorship. I'm arguing that we probably shouldn't. Our society is not really equipped to handle the free exchange of ideas at this point.

I'm naturally sympathetic to the sentiment, but this post fails to acknowledge, much less address, the major societal changes that have precipitated our current situation. Censorship is a reaction to our society's increasing inability to combat misinformation. Over the last 20 years, a huge, extremely influential right wing media establishment has sprung up. It has clearly demonstrated its ability to convince millions of people of things that are factually untrue, with less logic than repetition. (Ask Barack Obama, the Kenyan Muslim, about the utility of reasoning). I am worried about someone who makes an argument that gay people are evil, because it probably now means that a group of commentators have adopted that simple statement as a talking point. Furthermore, they've tacked on preposterous arguments, logical fallacies, and agreed to engage in Ad Hominem warfare against anyone who argues differently. It's been an extremely effective formula for persuading a huge portion, if not most of our society, into believing falsehoods.

Unfortunately, the John Stuart Mill marketplace of ideas is failing to distribute value as obviously as our economic marketplaces. The openness of communication made possible via technology has turned us all into Sophists. Efforts to confront and address this reality is peripheral at best. As long as this is true, censorship will be an appealing option to keep bad ideas from spreading. Is censorship worth the cost? I don't know, but I'd lean toward saying yes, while fully acknowledging that the cost is huge.

People would just argue about the rules and logic embedded in the tagging algorithm.

You're assuming that the capacity of the Powerwall is all the power available. With a solar-tied backup system, there is power coming in from the modules as well which is additive to the discharge rate of the Powerwall if wired correctly.

So even your massive A/C should be useable for a while during daylight hours if your solar system is large enough. Most people here in humid hurricane country opt to have a 15 amp 120v window unit available instead, and limit themselves to a single comfortable room.

Inverter failure is real, but the industry standard has risen to 12 years for string inverters, and 20-25 for micros. Even for string inverters, most of the residential/small commercial market has moved to using optimizers for voltage regulation, which means the inverter itself is a lot lighter weight/cheaper.

Backup batteries are still relatively rare, as they should be - they are fairly expensive as backup power. Most systems remain grid tied. Smaller systems don't require retail net metering for a solid payback, since almost all of the energy is used as it is produced.

Roof repair is again, real, but not as likely as you might think. Asphalt shingles under an array are protected from wind, and even more importantly, heat. They degrade at half the rate of the rest of the roof. Assuming 12-15 years for roof replacement, you will hit the module warrantied life of ~25 years before you need to replace the roof underneath the array.

Cleaning the panels really only comes into play when the pitch of the roof is relatively shallow (3 on 12 or shallower). Otherwise rain will clean them fairly well. A lot of people won't have any trouble walking such a shallow pitched roof, and will simply hose the panels off once in a while.

Also, in general, the price of modules per watt is about half of what it was 5 years ago. Peripherals are a major component of the improved economics, but module prices themselves have mattered just as much.

Leave it to a #slatepitch to explain why Snowden would choose Hong Kong:

http://www.slate.com/blogs/the_slatest/2013/06/10/hong_kong_...

To expand on this scenario: Snowden knows that the bureaucratic wrangling to get him out of Hong Kong by the book would be significant. He assumes he'll get at least several months of effective asylum in Hong Kong while having easy access to high speed internet, professional filmmakers, etc., all of which are crucial for waging an international public relations campaign. Meanwhile, he's geographically adjacent to China, home of the only government in a strong enough global position to openly vie for him to release state secrets in exchange for protection. This puts the U.S. gov't into a difficult position: they can extract/assassinate him illegally, which of course is a public relations nightmare. They can wait six months to extract him legally, and take the chance that he'll make a deal with China (or be illegally extracted by the Chinese). Or they can look the other way while he pursues another option - asylum in a country like Ecuador for instance. That last option is his only real hope. Basically, that he's able to put enough pressure on the U.S. gov't through public relations and international positioning that they don't pursue backdoor sanctions against a country granting asylum.