"24/7 carbon-free energy" is actually a much stronger goal, because (for example) it's physically impossible to buy solar/wind energy on a calm night, so they would need to pay more for an alternative like batteries or nuclear.

"In 2024, we increased our 24/7 CFE percentage from 64% to 66%, and nine out of 20 grid regions with Google-owned and -operated data centers achieved at least 80% CFE."

Look at those numbers. 100% still is a very ambitious goal. Currently 0 global customers are buying 24/7 carbon free power from the grid, and even getting from 0 to 1 will require a lot of boots on the ground building real infrastructure. Afterwards, getting from 1 to [everyone] is just a matter of scale.

Even the "lazy" strategy of buying up capacity from existing nuclear plants requires outbidding the existing customers (if the law allows it), and only works in regions where nuclear plants actually exist.

> It's going to happen anyway for at least a large portion of every day simply because it's cheaper.

That's precisely why a 24/7 goal matters. Energy is already cheap and clean for a large portion of the day, but covering the rest of the day will require significant investment.

"Carbon offsetting is like saying to your husband or spouse: ‘I’m going to cheat, but I’ll buy you a diamond ring.’"

Sometimes, the carbon offsets are from projects that would be built anyway, and there is a lot of sketchiness all around.

It might be better than nothing or it might empower companies to pollute more and then say, "Look, we are offsetting all this!"

Some of these offsets are of dubious quality. E.g., I think some of these offsets are from _not_ doing something that you might have otherwise done, but maybe you wouldn't have done that thing regardless.

The biggest problem with carbon offsets is that they're not pinned to a particular time of day.

If you buy a 1 GWh block of wind/solar energy for a month, you're not doing anything to supply clean energy when it's dark and the wind isn't blowing.

"24/7 carbon free" means increasing the time resolution from 1 month to 1 hour, and funding projects that deliver clean energy when you're actually using it. This includes stuff like batteries, nuclear, and geothermal.

That assumes the seller of the offset actually offset something rather than created an opaque paper trail that makes it impossible to know if they did anything at all. I don’t think Carbon market makers have a good track record.

> Electrical signal attenuation increases with the square of the distance

not true. in standard HV-AC lines, power losses are ~10% per megameter. HVDC gets to 3-5%. So Nevada to Seattle would be at most 20% loss, and in practice 15%, and with HVDC closer to 7%.

https://www.nationalgrid.com/sites/default/files/documents/1...

> Electrical signal attenuation increases with the square of the distance

Power transmission lines at 60Hz primarily have ohmic losses, which are linear with length of the conductor.

Interesting fact - Power transmission lines are long enough that the capacitive and inductive effects do matter a little bit, even though it's only 60Hz. That's why spacing between conductors is important. 3-phase lines will also rotate the order of conductors every so often to keep the average spacing between all pairs of lines similar.

I think you're over estimating losses for high voltage transmission lines. It's "only" 800 miles from the Mojave to seattle. In China there is a high voltage transmission line over 2000 miles long

> As a rule, if you're tempted to begin any sentence with "they should just..." -- don't.

Strongly agreed.

> Electrical signal attenuation increases with the square of the distance, so you'll lose ~95% of the power to heat loss in the wires if you try to power Seattle from solar in Nevada

What? HVDC lines are usually estimated to have 3.5% power loss per 1000 km. Since power transmission is done using power lines, the inverse square law doesn't really apply here.

> There are a lot of very good reasons why we haven't covered the desert in solar panels.

That does remain true however. Cost concerns, grid access concerns, environmental concerns are all good reasons.

Confidently, impressively wrong. Imagine how the power grid would work if there were 95% losses over 900 km.

Seattle is fine on power, you can pave the parts of the Mojave in California to power California instead (although tbh I think what we need most in CA is storage for when the sun goes down).

I think you can also reduce heat loss by cranking the voltage up, right? I imagine that's how current interstate/cross-country power deals work

I'll ignore your other points, because sibling comments are addressing them better. So here's something more unique:

> Also the permitting of stuff across state lines is so time-consuming it's effectively illegal.

This is true in general, but in this specific case, there are a lot of obvious ways to get around the problem, because Nevada is a moth-eaten shirt of federal land reservations — Nevada-the-political-entity only owns/regulates ~15% of the land of Nevada-the-geographic-territory.

With the current state of the US federal government, lobbying to privately use one of those federal reservations would be a walk in the park; and once you're going "California -> federal land" instead of "Calfornia -> Nevada", regulation gets a lot simpler.

Fun fact: there's a National Forest in Nye County (bordering California) that runs right up to the edge of the DoE-reserved area where they did the nuke tests. The feds are fine with running HVDC lines through National Forests (they're not Parks, after all), and "repurposing nuked ground for solar" is actually an easy-to-sell narrative at all levels. You could build solar there and backhaul it to California without ever touching land regulated by Nevada-the-political-entity.

Seattle isn't a good example, as they have been carbon neutral since 2005. They have lots of hydro power in the Pacific Northwest.

1% from a Dyson Sphere would be a better energy return than 100% efficiency from any conventional energy sorce. Similarly, 20% from solar is competitive with 60% from nuclear, 50% from coal, and is easily better than 100% from my neighbor Dan riding a bicycle powered electricity generator.

You can't cite efficiency percentages in a vacuum to imply they are a better or worse than alternatives, because those aren't percentages of the same kinds of things, and they don't tell you about the economics, production in absolute terms or EROEI.

They're blue, and nothing about coal would be less impact.

Coal starts with pulling down entire mountains to get to the coal. The whole process starts with environmental destruction and that's how it ends.

The thermal mass of the panels is no where near significant. Especially compared to a run away greenhouse effect we know coal to cause.

Definitely not. At global scale, the offset effects of solar installations outweigh albedo effects on the order of about 30x. [0][1]

  [0] https://www.nature.com/articles/s43247-024-01619-w.pdf
  [1] https://acs.figshare.com/articles/journal_contribution/Net_Radiative_Forcing_from_Widespread_Deployment_of_Photovoltaics/2871685

> * Carbon free energy is cheaper to produce these days, so the premis is wrong

If it's cheaper to produce, then there's no point in doing anything special, because it'll just take over via market forces. The "problem" with renewables is that the economics are fantastic in the marginal case, but if you need it to take over baseload when the sun is down, wind isn't blowing, etc., it's much more expensive because you need to build energy storage, which fossil fuel generators get for free. Going from 50% of power generation to 100% is going to be many multiples more expensive than 0% to 50%.

Until things like flow batteries become economical, in places without rivers you can dam for generation/pumped storage, it's probably not going to be feasible to go 0% carbon unless you just replace all of your baseload generation with nuclear or something.

This is absolutely true. The strategic goal of "dumping" is to drive competitors (the example here being the US solar panel industry) out of a market, with the implied goal of raising prices at the end of the process. But it almost never works that way in practice.

Pretty much by definition, products that can be dumped are ones that can be manufactured with easily scaled production equipment at low margin and high volume. If you can dump your competitors out of business, so can someone else, to you.

And that's the way it generally happens (c.f. DRAM in the 80's). The "dumping" phase of a technology curve is succeeded not by a cartel with a tight grip on the previously competitive market, but by an extremely competitive market of producers all undercutting each other. Almost everyone wins, except the original high-margin producers. But they (c.f. Intel in the 90's) tend to move on to other products anyway. No one wants to make junk for a living.

If you’re really smart, you exploit the cheap solar panels to provide energy for your own solar panel factories.

> at the cost of all potential future industries.

And at the cost of the future, period. Climate change promises to be extremely onerous to future generations, from a purely economical POV.

Thank you, jesus. Finally someone who understands the dynamics involved here. If you want to build up / protect your domestic industry then there are way to do that, but methods which don't involve also snapping goods sold well below market are crazy to not do so.

okay. I thought it was insightful enough. I've worked in many economic unions where that mismatch existed and needed further investigation to steer organizations towards the desired outcome.

pointing out that the current efforts don't do that is all that needed to be said. your criticism of that comment, to me, proved that you understood the comment enough that one needs to look further into the motivations.

>There has been a move to get the FTC to start labeling these net carbon numbers as misleading advertising because it always includes a bunch of purchased offsets unrelated to the company

Which is good, because carbon offset are a scam.

Solar power is much more expensive if the power must adhere to some number of 9's of reliability.

Often data centers need to run at three nines of availability (ie: 99.9) and at that level of availability the price of solar power is many multiples of conventional power due to the cost of batteries and over provisioning.

Not just that either. Once the current installation has completed its life cycle, the replacement could come from any source - domestic, alternate or original. And since PVs generate DC supply, interfacing even significantly newer generation panels/equipment isn't going to be too hard.

remarkable

No alternatives have been tried on a global scale, so I don’t think anyone has the practical evidence you request.

Resource allocation? Yes, normal socioeconomics. But did it have rapid growth trajectory? No. It lacked the ambition needed to drive humanity to maximum consumerism.

Before capitalism and communism was a thing, we had normal economics. And the worst part of this in some places was the bourgeoisie. They kept the resources in their control. And communism was supposed to be a banging of the piñata that shook the tree. Capitalism was a response to communism, yet installed a basic transition to a socialist education system and for educational wealth to be added.

I guess a good antidote to bad actors is for a wealth tax. Unused or underutilized wealth gets taxed until the net tax is reduced to whatever financial return it has.

I'd wager the biological cellular economy is more efficient and more robust than capitalist economies.

Not that I advocate nature red in tooth and claw as the epitome of governance.

There are none. But I don't want the economic system to be judged by its ability to do resource allocation. I want my economic system to be judged by its ability to improve metrics of human well being - and not even the median well-being, but the well-being of say the 10% percentile person in society.

Sure, resource allocation can be a instrumental to those goals of human well being, but not the directly optimized metric. So, for example, I care very little gdp/capita when you can just as easily measure life expectancy or food stress levels.

On such counts, capitalism has a mixed record. Today, for example, a lot of countries that are "more capitalistic" than others have poorer metrics of human development than those that are "less capitalistic".

They're a dinosaur and have forgotten how to innovate and provide new value. This is what complacency and shifting your business to focus on regulatory capture ultimately leads to.

Shit, I hadn't had that thought but it's self evident, the "worse search = more queries". Thanks for the insight.

Like, uniquely among all other industries? You don't think the petrochemical industry, for example, needs to pay for its larger externalities?

Who passes those laws? The politicians bought and paid for by the same companies?

Unfortunately not all of us can establish a religion capable of organizing a continents worth of peasants into being alright with the status quo that keeps a boot on their neck.

I’m sure many aristocrats buried without a head in France also thought their money would buy security from mob justice and insulate them from the breakdown of the social hierarchy that protected them in the first place.