Hacker News new | past | comments | ask | show | jobs | submit login
Google R&D team's goal: Make renewables cheaper than coal "in a few years" (fastcompany.com)
79 points by elbrodeur on June 7, 2011 | hide | past | favorite | 51 comments



This stuff is the reason I'll never hate google the way I hate Microsoft. Ok, hate is a strong word, but I just don't like or trust microsoft. I think the world is a worse place because of it. I Think the world is a better place because of google, but there's still a healthy portion of sceptisim that comes with that feeling.

The reason I mention this is because right now, it's all about google becoming the next Microsoft, but I don't think that's fair on google.


I think your hatred of MS is misplaced. The world is almost certainly a better place because of them. Maybe not as better a place as possible, but certainly better. The "MS is evil" meme always struck me as an immature response to MS's imperfections, few companies have done as much to put computing power into the hands of people, and that has had an amazing impact on the world.

As far as google, they'll lose more and more of the fire from their early days as they get bigger and older, but they still have a fairly strong ethical core for any company their size and I think that will help as they age.


I don't think you are giving enough credit to the companies and services that never were (because of Microsoft). Infact an uncharitable reading of what you've written would be that you assume no microsoft means some kind of industry void.

It is certainly pretty disputable that computing adoption was in anyway accelerated by the exsistence of Microsoft since they have essentially been rent collecting and pushing up prices for everyone to join in.


Sure, it would have been different, but in how many of those parallel universes would it have been better without MS? Keep in mind that the biggest competition for MS in those early days was Apple and IBM. Apple of that era coming out on top would have resulted in far fewer people gaining access to computers and far, far slower rate of innovation, especially in hardware. The same would have been true for IBM coming out on top as well.

Microsoft was one of the few companies which pursued an open platform for computing hardware and a licensed OS. That very much accelerated innovation in ways that most of MS's competitors did not.


OS/2 was an open platform, and would have had more or less the same effect as Microsoft's dominance.


You have to draw a distinction between what IBM did as a response to competition in our actual history and what IBM would have done had Microsoft never existed. IBM would have almost certainly continued what it had done before and what virtually every other company was doing at the time: offering bundled hardware/software solutions, using customer lock-in and FUD like weapons, and pushing incredibly high profit markups on everything. It's certainly the same thing Apple did. And it's the same road that smartphones have for the most part taken up until very recently. It's easy to skip past the fact that Microsoft had a pretty significant impact on the structure of the entire computing industry and some of that impact was very positive. Granted, they certainly have their fair share of sins to atone for but in my judgment any of their likely replacements from the early era of personal computing would have had as many or more.


the companies and services that never were

BeOS. Sweet lord, what BeOS could have been...


Although I love BeOS, do look at who was running the company and what that same individual did to Apple. The amazingly bad moves made by Microsoft's competitors sure helped a lot. Much like Microsoft's bad moves are helping their competitors now.


BeOS seemed like a very cool experiment but from what I've heard it was sunk by its own faults, most especially from being nearly impossible to develop for.


Hardly. I was a part of the developer program, and for the time, it was a dream to develop for. Because of the pervasive multithreading, development was a little bit more complicated, but the BeAPI was lovely, and very forward-thinking. And the benefits of the multithreaded architecture were immediately apparent to anyone who did even elementary development.

What killed BeOS was Microsoft strong-arming PC manufacturers to not allow BeOS as an option. Be had Hitachi and Compaq lined up for dual booting and an internet appliance, respectively. Microsoft used their OEM program to get them to go back on the deals.

BeOS alone is one reason I will never forgive the anti-competitive, monopolistic tactics of 90's Microsoft.


Be (the company) offered to give BeOS away for free to any OEM, but none would touch it... because the contracts with Microsoft at the time meant they had to pay a Windows license for every machine sold, even if BeOS and (not Windows) was installed on it.

It wasn't a good time to compete with Microsoft.


That, not the whole Internet Explorer tangent, was the thing that ticked me off the most. The thought that any money for the Intel box I bought to run NeXTSTEP went to Microsoft was unbelievable. If the gov had put a stop to that one practice earlier, it would have been a different game.


MS was evil, at least earlier in the company's lifetime. The theft of IP from Gary Kildall, the Stac case, the Wang lawsuit, the way IBM got shafted over OS/2, etc.


For my money, the single greatest evil Microsoft has ever perpetuated is Internet Explorer. And they show no signs of giving up the ghost on that one. The World Wide Web would get my vote as mankind's single greatest achievement, and I think we're still right at the very beginning of it's chapter in history.

Sadly, my view is that we're 10 years behind where we should be because progress is anchored to Internet Explorer, and will be for another 10 years at least.

I will happily call IE evil because Microsoft have used it as a weapon in numerous attempts to seize control of something they didn't create, don't understand and above all have no right to mess with.


> The World Wide Web would get my vote as mankind's single greatest achievement,[…]

I'd place the Internet before the world wide web. The www makes great use of many-to-many communication, but the Internet enabled it (which was its entire point, I think).


It's true, there's no WWW without the Internet, but I don't think the Internet would be what it is today without the World Wide Web. Perhaps something else (something better?) may have come along in it's place, but it didn't.

I think the success of the web is it's ease of use, which has enabled anyone with an internet connection to contribute, and it's these contributions, both large and small, technical and non technical that make it mankinds greatest achievement.


It's PR. As Page admits, it's how people see Google from the outside, but actually only 3-5 people somewhere in the company. There's none of the security that you might expect from a big company doing it - for example see what happened to http://en.wikipedia.org/wiki/Unladen_Swallow The main way that Google helps the project is providing publicity. Which aligns with Google's needs. It's PR.

PR is a crucial part of Google's strategy, because it's very easy to switch search engines.


No Microsoft means no (or at least a very drastically different/less funded) Gates Foundation. No Gates Foundation means a world much worse off, particularly for those in the 3rd world.

Granted, Google is more directly involved in positive projects as a corporation than Microsoft has been, but "The world is a worse place because of Microsoft" is a very shallow view, and IMO very wrong.


If you really want to do the expected utility calculation, you'd need to ask if the absence of Microsoft and the Gate foundation would have permitted wealth to be created and moved elsewhere, perhaps to another foundation.

Anyway, if the goal is to slay Poverty, a private foundation probably won't cut it. We need to do away with the current fractional reserve banking system, nullify most national debts, localize production and distribution of common goods (especially food), build things to last (no programmed obsolescence), and stop thinking that GDP growth is a good measure of wealth (it was a good predictor, but now we tend to cheat).

But the amount of collaboration needed to do that looks so huge that it may be easier to build a Friendly AI.


The Bill & Belinda Gates Foundation != Microsoft. It's true that it wouldn't exist without Microsoft, but the two are entirely different enterprises, with entirely different characteristics.


The Gates Foundation does great work. They are certainly one of if not the most effective anti-poverty charities out there. But positing a world "much worse off" is stretching things. Even huge charities are dwarfed by the sizes of governments. Almost all aid is delivered by governments. How much money has the Gates Foundation spent in total over its decade or so of work? What is the USA humanitarian aid budget for a single year?

Again, my point isn't to diss Bill's charity work. But arguing that we should tolerate abusive monopolies in our tech industry because the owners might one day get rich and give the money to poor people is ... borderline insane, honestly.


How much money has the Gates Foundation spent in total over its decade or so of work? What is the USA humanitarian aid budget for a single year?

US federal Humanitarian aid budget for 2009: $4.3bn

Gates Foundation total charitable distribution 2009: $3.65bn

Gates Foundation total grant commitments since inception: $24.81bn

So it seems you're either seriously over estimating The US Government spending or seriously underestimating the Gates foundation.


That sounds like a straw man. He wasn't arguing for monopolies. He was just saying that The Bill and Melinda Gates Foundation should be added to the postitive effects which Microsoft has contributed to the world.


>Almost all aid is delivered by governments.

I'm pretty sure that the opposite is actually true. I don't have any stats on hand, but I've read before that private donors in the U.S. give more foreign aid than any government on Earth, including the U.S. government.


Getting solar this summer (live in MA which has very nice tax deals) and one of the interesting things I noticed in the process was how much of the cost is the install. Saw one company that is making the install process significantly cheaper which results in a better win overall than if the cost of the cells were dropped by half. Same goes with the idea of transparent cells. Even if they have a very low rating they can go in every window and get installed for "free" when the new windows are installed making the payback much faster.


Do you have a breakdown of your own costs? Even in percentiles that would be interesting.


Yah I did a full breakdown and worked out all the numbers* to calculate the time period till payback (4-6 years) after which there will be some end large profit from the tail end of the MA SREC's and then small but continuous "profit" from very little to no electric bill (depending on my AC usage). After all the work I did my numbers were very close to the quotes I got.

After the immediate state rebase + fed and state tax rebate I am only paying ~3cents/Watt

Any particular number you want to know? Say the system costs 24K, MA does a straight rebate to the installer so you would only have to write a check for say 18K, and come next tax year get 6K from the fed and 1K from MA back so the 24K system would cost you ~12K. The SRECS pay that back between 3-7 years and that is guaranteed $300/MW till 2020 so by doing the install this summer I should pay it off around 2018 and have extra income until 2020. Oh and it doesn't count against property tax for 20 or 40 years (forget which).

The main reason I am doing the switch is that it is a small investment that is _extremely_ stable with 1) a 100% return with interest in ~6 years on the electricity 2) increasing the value of the house by ~20K (more than post tax check!) and 3) will continue to generate electricity/tax free income for as long as I live in the house potentially the equivalent to the initial cost several times over. Not a bad small investment.

* Electric usage est, electric cost growth est (power costs more every year in MA), sun hours (obtained from various government sites) est, SREC payback est, power est, angle of house and roof.


Thanks for the detailed response. I'm surprised by how much of the cost is being deferred by subsidies, but maybe that will pay back in terms of spurring innovation.


2675 kWh system (about 3/4 of the yearly use of a regular family in the Netherlands) costs 6379 euros (9346 USD as of now) + 1498 euros (2194 USD) for installation. The system is sold as 'DIY' though so it should be possible to install it yourself.

Where I live there is a 1500 euro subsidy for the installation, putting the payback time (assuming constant energy costs) at around 7 years; with opportunity costs accounted at deposit account rates around 11.

I haven't decided yet but I think it's already an OK investment (not great but OK), mostly because of the subsidy; without that it would require significant increases in energy costs over the next couple of years to make it attractive.


I think Gladwell is wrong. Historically, there have been many cases where researchers have been unaware of relevant work, because it was done in a different field. Search could improve there. If search is the quest for knowledge, then the scope for improvement is unlimited.

Sinking to more consumer-oriented search, anticipating what you want before you know you want it is another (creepy) way for search to improve, that Google has specifically talked about.


The goal is very noble and I'm all for it, but there is this problem that the sun doesn't always shine and the wind doesn't always blow, and electricity isn't infinitely transportable, nor easily stored.


There are serious hurdles to decoupling ourselves from coal, but none of them are insurmountable. At least not hypothetically. :)

One interesting idea I've heard in regards to storing electricity is using electric cars as batteries. Granted, this would require a massive retooling of our grid and our transportation system -- not to mention the question of where we will find all the resources to make so many batteries -- but it's still interesting.

What's troubling to me is all the sensationalism surrounding nuclear, which seems to suffer from lots of near-religious objections. Nuclear is dangerous but even in the worst scenarios (like the Fukushima Daiichi plant) the damage is far less onerous than the continued operation of coal plants.


all the batteries that we have ever build combined cannot store 10 minutes of the electricity that we currently use. We need order of magnitude breakthrough with batteries first.


Usually a hydro power plant is used as a "battery". When the demand is low, the electricity produced by wind/solar power plants is used to pump the water uphill. When the demand is high, the hydro power plant releases that water to produce energy.

http://en.wikipedia.org/wiki/Pumped_hydro


That works in Norway, but would it work all over the globe?


They do it in Japan, and I think they do it in Quebec. Don't quote me on the second one.

They had to have rotating power outages for a week or two in Japan after the earthquake because of the need to replenish those reserves.


An article posted to HN some time back:

http://www.fastcompany.com/1708167/how-to-make-lithium-batte...

Discussion:

http://news.ycombinator.com/item?id=1986640

Summary: Infusing Li-based batteries with the Tobacco Mosaic virus could boost their storage capacity up to 10x. There's one order of magnitude...


And the tank is not always full, nor the hopper.

Requiring 100% replacement/utility is not a useful position. If we want to reduce (notice I didn't say eliminate) use of fossil fuels, then we have to pursue a mix (including fossil fuels).

Edit: You didn't actually require 100%, but this edit is an easier correction than a correction.


Interesting cadence to your comment, but the same could have been said of petroleum infrastructure before we had petroleum infrastructure.


Someone I know who runs a nonprofit related to helping remote fishing villages and the like in his Asian country, claims supercapacitors are less than 5 years away.

"If" that is true, then a time-shifting bank of such supercapacitors could presumably solve the very real issues you raise.


What do remote villages in Asia have to do with supercapacitors?


He is someone who works full time in researching alternative energy solutions for those fishing villages.

Wind or solar solutions that he specifies have to be rugged and easily maintained without access to fancy tools, and batteries can wear out quickly in high-temperature environments.

Thus he has been researching supercapacitors.


Supercapacitors are less than five years away, and have been that way for at least ten years now.


In the near term, the sun shines during peak usage hours. It doesn't need to be stored until it is a significant fraction of production.

The other no-storage option is to retain water at hydro facilities during the day and use more at night to balance the solar production.

That gives us plenty of time to see if the liquid metal battery will work out, or to start building reservoir pairs.


Somewhere the sun is shining and the wind is blowing. I can guarantee that Google will have at least one data center somewhere will one of these two cases will hold true.


I don't see much reliable wind, but the sun shines pretty regularly where I'm sitting... electricity is easily stored for short (12hr) periods.


We need cheaper battery storage more than anything I think.


The problem with this is always that they are "a few years away", or that we can run the entire economy from biofuels "in ten to twenty years".

It has been going on for what, 20 years now? It is never going to change until we ignore them until they actually bring about their ideas.


Yes, but in the last 30 years, we've went from prohibitively expensive 4.5% efficency cells in 1954 and $1500/watt in 1955 to a test 500kW installation in 1977 to $9.00 in 2007 to, in an ideal situation, 8.5 cents per KWH today.

http://inventors.about.com/od/timelines/a/Photovoltaics.htm http://greenecon.net/understanding-the-cost-of-solar-energy/... http://www.reddit.com/r/Frugal/comments/hn7jn/while_were_on_...

We could start building a 100% solar/nuclear solution in the US today. The technology is there, and it would only be the construction time. Our energy costs would go up, but not significantly. However, if the technology keeps on this curve, unsubsidized solar will be cheaper than coal soon.

We aren't doing this today because the power companies know that the curve is coming soon. Why invest prematurely, outside basic research projects?


Upvoted. Same with Wind. Learning curves are bringing costs down below that of coal and nuclear.

Solar is already in plenty of niches - from highway signage to parking kiosks. Every niche helps increase production, which reduces price per unit.

This is why I support subsidies - unlike for oil or nuclear, they're actually working to get costs down.


It's expensive to make things cheap. Usually when something becomes cheap quickly it's because demand is high and producing that thing is less expensive. In the case of new energy sources, reaching parity with the grid is very difficult: Not only are technologies like coal already operating at scale, they are also subsidized by the government.

If there was a strong public demand for renewables grid parity wouldn't be that far off. Unfortunately, we're a little too obsessed with other things at the moment, like politicians posting lewd photos online or which television shows to DVR.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: