Slashdot Mirror
Lessons Learned From Google's Green Energy Bust
the_newsbeagle writes In 2007, Google boldly declared a new initiative to invent a green energy technology that produced cheaper electricity than coal-fired power plants. Sure, energy researchers had been hammering at this task for decades, but Google hoped to figure it out in a few years. They didn't. Instead, Google admitted defeat and shut down the project in 2011. In a admirable twist, however, two of the project's engineers then dedicated themselves to learning from the project's failure. What did it mean that one of the world's most ambitious and capable innovation companies couldn't invent a cheap renewable energy tech?
is that software people who are used to downloading programs and typing things to make pretty pictures appear have *no clue* about the complexity of the real, material world.
This is why Space Nutters are mostly programmers. Hence their delusional beliefs.
"What did it mean that one of the world's most ambitious and capable innovation companies couldn't invent a cheap renewable energy tech?"
Simple, solar and wind are not the way to go to make cheap, reliable, clean energy today.
It is an answer that a lot of people will not like but that is the simple truth.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
I think this speaks a lot about how companies and the population are increasingly thinking in rather short terms and how little respect the modern tech elite have for those who came before them. There seems to be this attitude that difficult problems are only unsolved because the 'wrong' people have looked at it and flush with arrogance for solving comparatively simple internet related ones they believe that they are smarter and thus will quickly tackle what those 'researchers' and 'old fogies' could not.
And when gratification is not instant, they move on.
I also see this, on a smaller but more insidious scale, in the almost pathological desire to not learn from the past developers have been fetishizing. Too often learning roots or old technologies 'taints' a person with 'old' ideas rather than teaching them lessons others have already learned so that they can move on from there. So many 'new' technologies that when the developers are asked 'ok, this is great, but how do you plan to address the issues that were encountered last time?' they just look at you blankly and claim this is new and innovative, or that you just don't understand.
Ok, got a bit off topic there ^_^
Big difference. There isn't really much that Google does well, especially taking the size of their coffers into account. Some things just need economies of scale to bring the price into an acceptable range. Others need actual innovation. Google does not do well with the latter.
It means you ain't smarter than the generations that went before you.
If you don't like the choices previous generations made, you first should figure out WHY they made those choices before deciding they were wrong.
When I think of the word 'bust' I picture (female) breasts, but then I am a heterosexual male.
I have seen - predominantly on Slashdot, obviously, but also elsewhere, a sort of naive technocrats (who are often also libertarians) believing that as soon as some technology is needed, the invisible hand of the market magically creates this technology so one only has to sit and wait for this magic solution to appear out of thin air. The more down-to-earth kind of these people even tried to explain this magic by telling that this process happens by throwing enough money at a problem.
Unfortunately - and TFA is a picture book example of this - reality doesn't work that way. Breakthroughs don't happen by magic, they happen by meticulous research and a shitload of small steps. Solutions don't suddenly appear just when they are needed, a long lead time of research is required. And sometimes this new technology never comes up at all.
"It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
The big news to me isn't that they weren't able to invent the tech, but their estimate that even if their most optimistic scenario had come true w.r.t. clean energy tech that it still wouldn't be enough to avoid the "really bad" scenario w.r.t. climate change (if you trust Hansen's models).
YET. They haven't found a cheap renewable energy tech YET. Coal & other prices will continue to rise, while their efficiencies are the highest they will get. Solar cell costs are plunging, while their efficiencies rise. I predict a collision, a market and a profit.
I come here for the love
So some software guys could not solve a really hard problem in the physical world better than the energy guys? And nobody saw that coming (shocked)
You can't "invent" cheaper tech--it only gets cheaper if you invest in mass-producing it. They gave up 3 years ago, and since then market forces have actually achieved price parity for renewables in a lot of the world. It wasn't any new "magic bullet" research that did it, but incremental improvements driven by economies of scale. Yes, government played a big role, but primarily as a driver of demand and investor in manufacturing.
The climate does not have time to wait for a new technology to be developed and go through the whole sequence of commercialization and commoditization. The solar panels, wind turbines and batteries we already have can do the job--and the more we build, the cheaper they get.. This is one place I wish market purists would get on board--put a price on carbon, and solutions will come out of the woodwork and plummet in price.
What did it mean that one of the world's most ambitious and capable innovation companies couldn't invent a cheap renewable energy tech?
Umm, nothing. Google has no special expertise in energy tech. This is WAY outside their core businesses where they have a proven competence. The notion that they would to solve the economic problem of renewable energy where everyone else had (so far) failed is somewhere between well intentioned altruism and pure undiluted hubris. (not sure where on that scale though) The only thing Google has is smart people and a huge bank account. Those are nice assets to work with but just because you can throw smart people and money at a problem doesn't mean a solution will magically appear in a timely manner. Research is unpredictable and requires long term dedication. And even if you do succeed in coming up with a nifty new technology it doesn't automatically mean that the economics of it will be favorable. I'm not saying Google shouldn't try - I'm glad to see them working on and/or bankrolling research such problems. My point is that Google shouldn't be expected to be more likely to solve the problem than any number of other companies/organizations that have worked on these problems.
Yes, it's not renewable.
Yes, it's petroleum based.
Yes, it's not clean.
But it is cheap, reliable, commercial off-the-shelf technology that is far cleaner than coal.
What it means is that Google has a tendency to assume the set of intelligent people in the world (outside academia perhaps) is a subset of the set of Google employees.
If you don't like the choices previous generations made, you first should figure out WHY they made those choices before deciding they were wrong.
Most of the time the answer boils down to "it seemed like a good idea at the time". We use fossil fuels because they were available and we figured out how to make the economical sooner than some of the alternatives. We didn't know about some of the environmental side effects at the time. Same with nuclear. We tried all sorts of things with radiation that we now consider insane because we didn't know any better at the time. We figured much of it out in time but we didn't magically know all the problems with a technology the moment it was invented. So we build on what we know at the time and sometimes (like with fossil fuels) find out later on that maybe what seemed like a good idea before really wasn't. That's ok. What's not ok is doing nothing once you realize there is a problem.
You are wrong. The gigantic economic boom of the 20th century was fuelled by cheap oil.
There is no single cause for such a complicated occurrence. Oil is certainly a part of the equation but there are a LOT more variables than just the price of oil.
It is, and will continue to be, so long as governemnts keep paying people to not build fusion reactors.
Please cite a single incidence of any government actually paying someone to not build a working fusion reactor.
My guess is that, when we finally get a working fusion reactor, it will be developed in a few years by a company that completely ignores all the 'basic research' governments have funded over the last fifty years.
Based on what? Something more than a hunch I hope. Or perhaps the simpler answer is that it's a really tough problem to figure out. Research doesn't care who funds it. Either the findings are useful or they aren't.
It definitely creates earthquakes reliably.
Solar cell costs are plunging, while their efficiencies rise. I predict a collision, a market and a profit.
You might see one, if you could just plug solar cells into your house and magically get power all day. Most of our power usage in our house is at night, when... oops... there's no solar power.
No, actually, in America the highest electrical usage is in the afternoon. It's driven by air conditioning loads in summer, along with the fact that business and industry tends to use the most power only during working hours. There's a slight bump at about 7, but it's not as big as the afternoon peak.
Quick calculations suggest that you can replace about 10% of US electrical usage with solar with no disruption at all, and something like 20 to 30 percent with only minimal disruption.
That's not enough to solve the energy problem. But, with the electricity market in the US at something like half a trillion dollars a year, that's a substantial market (and substantial profit)
http://www.geoffreylandis.com
Solar cells could cost $0, and they still probably wouldn't make sense when compared to grid power that isn't made artifiicially expensive by Greenist boondoggles.
You might have a point if fossil fuel generation actually had to pay for all the environmental damage it causes. But since they don't and the real cost of heating your home is higher than you might guess from your monthly bill.
And if you think solar cells for free would not make economic sense for a huge portion of the population then you have NO idea what you are talking about. Nothing actually costs zero but super cheap solar cells with good efficiency would massively change the world energy markets.
So yes, the bulk of our power usage (and Im not the poster you replied to) is over night.
So of course your usage habits clearly apply to everyone else in the world and nobody is ever in their home during the day. [/sarcasm]
If you want to trade anecdotal evidence the bulk of our electricity usage is during the day during the summer when our AC is running. Most of the night usage could easily be stored in a battery bank that could fit inside our house.
As opposed to your delusional beliefs that Google is just "software people", that "software people" have no clue about the real world, and that "Space Nutters" are mostly programmers and delusional, and that labeling people and categorizing them into overly-broad buckets makes you look intelligent.
The problem is that is only true for a few hours around solar noon.
For now. Furthermore a LOT of energy usage occurs exactly during that time. Let's say you have a grocery store and you need to power refrigerators. Guess when the maximum power drain from air conditioning will be? Exactly during mid day when the solar cells are at maximum efficiency. In my opinion most industrial businesses should have a rooftop covered in solar cells. It is wasted space now, it generates clean(er) energy right when it is needed, it distributes the grid, turns variable power costs into fixed costs and it should reduce costs in the long run. Rooftop solar on businesses makes a ton of sense. Doesn't solve every problem but it would be a big step in the right direction.
Just because it doesn't solve every problem doesn't mean you should dismiss the problems it does solve.
Wind is much better but people always seem to ignore they production vs consumption cycle problem with solar.
That depends very much on where and when you need the power generation. Wind is great but like solar it solves some but not all problems. It's not hard to find use cases where solar makes more sense and others where wind is the better option.
(The 1st AC is right).
Watch this Heartland Institute video
The more down-to-earth kind of these people even tried to explain this magic by telling that this process happens by throwing enough money at a problem.
New "magic" technology is not really needed at this point. What is needed is a policy shift. As long as externalities are free for some but not others, one technology will be cheaper than the other.
Case and point is nuclear power vs. coal/gas. Both are local employers, thus do not affect trade balances (generally, at least until not coal). But one is required to deal with ALL its waste, while the other is allowed to pollute CO2 without consequences.
Another example is fossil fuels vs. small scale renewables and "balance billing". Fossil fuel (or power plants in general) sometimes get negative grid price (they have to *pay*) to send power to the grid, while small renewable installers get net billing even when their power is not needed. This is a very bad distortion of the energy market.
So there you have it. Two very good examples where policy dictates how external costs are or are not dealt with. And it is this policy that distorts the market prices. Policy can be used for good effects (like revenue neutral carbon tax, academic research funding), or bad effects (no price for CO2 pollution, net billing schemes, subsidizing specific technologies, etc.)
Then again we supposedly elect people to make decisions but think we know better anyway. This causes lowest-common-denominator policies to be implemented. And in case of Global Warming, that means nothing gets implemented as the wealthy can just skew the thinking of the stupid voters enough to block any policy makers from even thinking of making correct policy decisions. In effect, the current status is exactly what we deserve, not what we need.
that outside of sci-fi, and despite what the "creatives" have claimed for centuries, that objective reality has limits and there are just some things that cannot be done without trade offs. There is yet no magic power source that is completely clean, lasts forever and has infinite power density - the one thing that all future near utopias seem to require. Star Trek needs it's "dilithium" to power that shiny happy Federation. Otherwise the shields that allow a purely defensive posture don't work, the replicators that feed everyone fairly don't function, and the weather control that keeps people from freezing or broiling to death in adverse climates doesn't work.
We have limits on what we can accomplish in this universe, and until new theories come out, those we know of are HARD limits that no amount of wishing, "thinking outside the box", or politically-motivated propaganda can overcome.
Imagination in engineering and invention is NOT the upper limit. I think in most respects our imagination has already explored beyond what reality can offer.
Time to start thinking ABOUT the box rather than outside it. Invest in the most efficient solutions with the least amount of negative consequences. And stop thinking that tradeoffs can be ignored or do not exist.
Perfection does not exist in this reality. Antitheists remind us of this on Slashdot every chance they get. But they need to also examine their own "faith" in application to practical needs and the belief that imagination will trump physics.
dunkelfalke (91624) writes:
I have seen - predominantly on Slashdot, obviously, but also elsewhere, a sort of naive technocrats (who are often also libertarians) believing that as soon as some technology is needed, the invisible hand of the market magically creates this technology so one only has to sit and wait for this magic solution to appear out of thin air. The more down-to-earth kind of these people even tried to explain this magic by telling that this process happens by throwing enough money at a problem.
--and Anonymous Coward responds
I have seen - predominantly on Slashdot, obviously, but also elsewhere, a sort of naive technocrats (who are often also liberals or progressives or socialists) believing that as soon as some technology is needed, the state magically creates this technology so one only has to sit and wait for this magic solution to appear out of thin air. The more down-to-earth kind of these people even tried to explain this magic by telling that this process happens by throwing enough money at a problem.
OK, somebody should moderate both of these as "troll".
There is some insight here, but the insight is completely washed out by the gratuitous insults and use of deliberately slanted vocabulary.
In fact, the market is good at solving some types of problems. And government is good at solving some of the types of problems that the market isn't good at. But people of all political views always call approaches that don't fit their ideology "throwing money at the problem." If it's a solution that fits your politics, it's "investing in technology," and if it's a solution that doesn't fit your politics, it suddenly "throwing money at the problem." Same thing, different choice of spin.
But randomly insulting political positions for the joy of insults, and substituting buzz words for thinking, really does not substitute for actual analysis.
http://www.geoffreylandis.com
Or do you merkins have some weird shit with your electronic white goods where they won't work in the daylight hours?
Not one sentence about lowering demand for energy?
and no amount of fiddling by a hack Supreme Court can change it. not even Google.
if this is supposed to be a new economy, how come they still want my old fashioned money?
You can't "invent" cheaper tech--it only gets cheaper if you invest in mass-producing it.
That's only sort of half true. You can invent a technology that is fundamentally less expensive than previous alternatives at the same production scales. It's not merely a matter of more = cheaper.
Lots of smart people + huge bank account + significant access to the world's accumulated knowledge.
None of which is unique to Google. Google does not have privileged access to more than a tiny tiny subset of the world's information and none of that is specific to energy technology.
It's a formula worth trying on the problems that matter - in the past, lots of smart people and a huge bank account have only been applied by governments to military applications.
Baloney. AT&T, IBM, Westinghouse, GE and lots of other companies have had huge bank accounts and Nobel prize winning research departments. Bell Labs alone was responsible for 8 Nobel prizes, unix, C, transistors, lasers, CCDs, radio astronomy and more.
Computer guys have a very dangerous tendency to think that because they can fix Mom's computer, and people are always asking for their help, that they are somehow much sharper than the regular person. All it really means is that they have some specific information that others lack. This leads to the absolutely sickening arrogance you see exhibited here all the time,
Actually, in Star Trek, the dilithium is just needed for the warp drive. (Notice that the dilithium crystal matrix is part of the *warp core*.)
I did read the article, which was filled with all of the appropriate doom and apocalyptic visions, but the ultimate conclusion is really rather useless - hope and pray that magic comes along.
"Our society needs to fund scientists and engineers to propose and test new ideas, fail quickly, and share what they learn."
"We’re not trying to predict the winning technology here, but its cost needs to be vastly lower than that of fossil energy systems."
Simply *wanting* a technological innovation doesn't make it happen. Even massively funding all kinds of R&D doesn't necessarily make it happen - not all R&D is created equally, and unless you can discern between useful work, and not useful work, you're looking at huge amounts of waste.
> When your meter shows 0 net power, you have generated all the power you need (just not at the right time).
With zero net power, your electric bill is zero. Yay!
You tell your neighbor about it. He does the same, and pretty soon everyone is getting all of their power for free. Everyone generates power during rhe day when they aren't home, and uses power at night, when solar doesn't work. Nobody pays for anything.
Of course since nobody pays, there's no money to generate power at night. Moral of story - solar-electric can work, but only if nobody but you does it. If you tell other people about there won't be anyone paying to generate your evening electricity.
...that Liberals DON'T have all the answers.
You know there is a really simply reason renewable energy is more expensive (except hydro and geothermal in favourable locations).
It's the second Law of Thermodynamics. Solar and Wind power is diffuse. Hydrocarbons and particularly nuclear are far more concentrated, thus much easier and cheaper to draw power from. If Google had invested in a array of advanced Nuclear Power technologies, one or more of them may have come off and we'd have cheap CO2 free power for millions of years. If may still happen but it is very difficult and the sophisticated simulations of advanced nuclear IS something where Google could really contribute.
Oh well,
Except market purists balk at this because "putting a price on carbon" is an artificial thing - it's screwing around with the markets. The markets have already spoken: the externalities of climate change (relocation costs, war, health costs) have a lower cost than trying to develop alternatives. These costs are already really accounted for, even though they aren't necessarily applied at the source of "carbon" emission.
No they are not. You completely miss the point of what an externality is in economics. The whole point of externalities is that they are *not* accounted for by market forces. If A and B have to decide whether to make a transaction, while C will be harmed if the transaction happens but has no say in whether it happens, that's an externality and market forces do not account for it under any economic model I've ever heard of.
If reducing carbon emission is a goal, pretty much all economists agree that a carbon price is the most market-efficient way of doing that, because it makes market actors make the most efficient decision while taking into account the externality, without favoring or penalizing any specific technology.
What starts out as a "sin tax" in reality becomes a method of wealth redistribution. Revenue from a CO2 tax would never in billion year directly fund "green energy". God, if only the tax system and Congress worked that way. Yeah, hell no it doesn't!
There's no reason a carbon tax has to go fund green energy. You could have a carbon tax that is revenue neutral because other taxes are reduced to compensate for it. The point isn't to fund green energy, but to price in the externality, and make each C02 emitter pay for the overall harm that CO2 causes. This in turn makes less-polluting alternative energy sources more competitive (thereby funding green energy through market forces) as well as promoting energy efficiency, without favoring any specific technology. This would lead to the most efficient ways of reducing C02 emissions "winning", as opposed to subsidies to specific forms of green energy which may end up promoting dead end or overly expensive technologies.
Dry rock (gravel) can store energy at about $1/kWh. It doesn't get any cheaper than that. How you use that storage capacity is to use a solar concentrator to heat a working fluid (usually water). Some of the steam goes directly to turbines to make electricity. The remaining hot water or steam goes to a heat exchanger, and a fan circulates air through that and the rock bed. When the Sun isn't shining, you reverse the fan, and suck heat out of the rocks to heat water/make steam again.
To keep the rock thermal bed hot, you surround it with "vacuum powder insulation", which has about six times lower thermal conductivity than fiberglass. On a large enough bed, that will stay hot for days. That's because thermal loss goes as the square of the thermal bed size (area), but heat storage goes as the cube (volume). So the bigger it gets, the longer it can store heat.
Batteries are not really a solution for storing power grid amounts of energy. A valley full of water (hydroelectric) or the equivalent of a gravel pit full of rocks (thermal storage) are answers because the raw materials (water or gravel) are really really cheap.
it just means they didn't have the right people doing the project....... And having money doesn't mean you can 'invent' something. Sometimes it's just the pennyless loner that invents the holy grail, not a well funded science team.. A lot of times those teams are held back by their fixed thinking...
I would think software developers would remember that in a complex problem there is no silver bullet solution.
Remember that coal is extremely cheap if you ignore the externalities.
I hope they didn't think they would come up with cheap and efficient solar panels that I can paint onto my roof plus the perfect energy storage medium that fits in my closet or even see a clear path to these things in a few years. This is a complex problem that would take some technology/engineering and basic science, but also a lot of political and economic thinking.
Did they even bother to look into the geo-air idea? Warmth in winter and cool in summer.
http://citrusinthesnow.com/
Perhaps something so simple is beneath them? Perhaps they were not aware of it. About $500 a year to run the fan (basically that is what it is) to heat and cool a greenhouse that is 80 x 15 foot greenhouse.
The problem is when you swing for the fences you strike out a lot. Just get on base first and see where that takes you.
This simply proves the thermodynamics of Green Energy is fail, as it has always been and will always be. That does not mean we shouldn't do Green Energy however: we actually don't have a choice, whether it's because of global warming, peak oil or economic collapse: it's not actually a choice for countries like the US despite what some people might want to delusionally believe.
We already had the cheapest and more thermodynamically efficient energy source: Oil and its petro relatives.
These have/had the best efficiency because of the basic foundation of thermodynamics: efficiency is related to energy reservoir to energy sink differentials. The thermal energy of sunlight to ambient 300K or air pressure differences for wind turbines are always going to be smaller (less efficient) than high temperature differentials inside of a gasoline ICE, kerosine jet turbine, or a coal or nuclear steam power plant.
Now the really bad news: we can't have "nice things" like oil or gas or coal forever. In fact, because of having reached Peak Oil already, we will have fewer and fewer "nice things". The reason is that the thermodynamics of all the alternatives are generally inferior thus the costs will always be higher. Lower efficiency means we get less from any energy we extract and thus we have less surplus for growth or luxury or discretionary uses. That's just how the future is going to be. Thermodynamics dictates this as surely as an apple falls because of gravity. It's not subject to debate at all.
Let's drive this to extremes just to see what happens. What if we manage to develop solar panels(and associated systems) that are 4X more efficient(they look black because they aren't reflecting light at all), and cost 1/10th as much. $20 gets you a panel that produces 1kw if given enough sunlight. By the way, at $20 per ~1.7 square meters, it would possibly be cheaper to build your roof out of solar panels rather than traditional methods such as asphalt shingles.
The practical result would be that electricity prices would invert - cheaper during the day. It might even be economical to use battery backup to not purchase(as much) power at night. From a practicality standpoint I picture individual homes using old EV batteries to provide power when it's dark out, with big power companies having more options. It would still be a net positive though - the cost of aluminum products would drop, electric heating would be back in vogue*, etc...
*Just that you'd have a big water heater tank that doesn't power on unless the solar panels have juice.
I don't read AC A human right
I had the same thoughts and I think it can be scaled up to store energy for an entire winter. One would basically excavate large holes in the ground and fit them with isolation for the store.
There are plenty of ways of transferring the heat into the store. The higher temp you use, the better (->Carnot's law).
It could also be used to store excess electric energy from ANY soure and release said energy by means of some sort of heat->energy thermodynamic machine. E.g. using helium or N2 to drive a gas turbine which would in turn drive an electric generator.
You would heat this thing like a glowing lamp (using Tungsten wires) and use a material with extreme high melting point (e.g. some Tungsten Carbide). That way you can achieve very good Carnot efficiencies up to 90%. You need exotic materials in your gas turbines, of course.
http://de.wikipedia.org/wiki/Wolframcarbid
Energy density for deltaT=2000K would be about 300J/kg, which is vastly better than pumped storage. At least if you are area-constrained like Germany and countries with simlar population densities.
Pumped storage has W=m*g*h = 1kg*9,81*1000m=10kJ/kg for a rather enormous system of 1km height.
But yeah, better pursue this kind of thing than "develop new arms to bomb the crap out of arabia".
The "nuclear" thing is a very clear example of the Power Of Propaganda. You can shit into people's brains very, very effectively if you do it long enough.
Irrationality 2.0