4 ms·
It is amazing to me that on the day Google announces they're going into the power business, the top story is an xkcd cartoon. Electric power is really fundamen
by pg 19y ago
It is amazing to me that on the day Google announces they're going into the power business, the top story is an xkcd cartoon. Electric power is really fundamental. Change how it's generated, and you change a lot. And Google could make a real difference. They have brains and money and a willingness to do new things. And (equally important) they can be their own beta user.
I grew up in this world; my father designed nuclear reactors. Hidebound doesn't begin to describe the power business. It's not (just) a conspiracy by a few big, classically evil companies. There is a whole culture of inertia. It is very exciting to think of Google doing a cannonball into that particular swimmming pool.
This announcement also shows how startlingly ambitious Google's plans are. Could you imagine Microsoft or Oracle, even at the peak of their founders' energy, doing something like this?
- rms 19y agoIt was a pretty good xkcd cartoon, you've gotta admit.
- inovica 19y agoAgreed. Shake it up Google!
- ivankirigin 19y agothe top story is an xkcd cartoon. I predict in 10 years, Google will mainly be known as a company that identified a flaw in social news sites :)
- andreyf 19y agoI predict in 10 years, Google will mainly be known... I predict xkcd will be.
- mxh 19y agoOn the one hand: GOOG is trying to build renewable that actually makes economic sense. Good for them. (I assume 'cheaper' means 'cheaper', not "cheaper when you factor in tax credits, gov't incentives, Renewable Portfolio Standards, RGGI, AB32, and so on". If it means the latter, this is a non-story.) Most renewable is a complicated ratepayer swindle. On the other: 2007 wholesale electric power transactions (outside of Texas) look to be around $270B. So GOOGs investment comes to (very roughly) 0.1% of the market. Describing this as a 'cannonball' seems .... dramatic. The industry is pretty hidebound, particularly on the T&D side. However, power is sold on relatively transparent spot markets in most of the country's ISOs (e.g. CAISO, PJM), and if generation technologies were sitting out there with significantly lower costs than what's in use today (renewable or no) I find it difficult to see how they could not have yet been deployed by some profit-seeking entrepreneur. I-banks play heavily in this sector, so it's not as if, were the numbers to make sense, it would be hard to find someone with the capital to fund clever ideas. I don't object to GOOGs actions: It's their money to spend, they sound like they're trying to come up with something innovative (though see my definition of 'cheap' above) and if they succeed, it will benefit everyone. I just think the odds are that the problem is a bit harder than they think, and that we're seeing a second prominent example of (slightly arrogant) SV s/w guys getting in over their heads. (Tesla is the first.) Still, good luck to them.
- euccastro 19y agoWorking with RE<C, Google.org will make strategic investments and grants that demonstrate a path toward producing energy at an unsubsidized cost below that of coal-fired power plants. I don't know if this covers your concerns about ratepayer swindles?
- mxh 19y agoGood catch; I didn't read the PR that carefully. 'Unsubsidized' is indeed promising - it seems to preclude direct kickbacks. It leaves open the possibility that Coal could be made deliberately uneconomic (e.g. C02 caps), and that the GOOG energy might be made economic through the sale of ancillary products (Renewable Energy Certificates) for which gov't creates an artificial demand. At this point, though, I'm just being suspicious. I suppose GOOG deserves the benefit of the doubt, and it costs nothing to give it to them. We'll see what, if anything, they come up with.
- raghus 19y agoIt's this startling ambition that I admire most about Google. They're trying to shake up everything from print/tv/radio advertising to mobile phones to power to ... Someday, we might all nod sagely and say "they should never have strayed from Search and AdSense" because it is entirely possible that they'll bleed a lot of money and fail on several of these - but it is great that they are audacious enough to try.
- far33d 19y agoI wonder what percentage of GOOG's operating costs are power related (heating, cooling, electricity, etc). I imagine that it's probably their highest expense outside of salaries. Cheaper and cleaner power (generated independently off-grid) could help them solve serious issues w/ remote data centers and their operating margins.
- ivankirigin 19y agoOReilly: One of the first times I met Larry Page, long before Google went public, when he was still driving a bright blue VW bug, he was telling me how for Google, the annual energy cost of each processor they put online was greater than the cost of the processor itself. (This may actually have been lifetime energy cost. It's a long time and I don't remember exactly, so I don't want to put words in his mouth.) Energy is one of the key costs for anyone running a large server farm, and keeping that cost under control is a key element of competitive advantage. http://radar.oreilly.com/archives/2007/11/google_energy_strategy.html http://radar.oreilly.com/archives/2007/11/google_energy_stra...
- far33d 19y agoThis becomes more and more true as your processors are more spread out all over the world, sometimes in remote locations. I heard google chooses data center locations based on proximity and access to power, which is probably very limiting.
- jey 19y agoThe trouble is that it's still vaporware at this stage. I'm very glad that Google, with its brains & money, is going into the power business -- but I'm not going to get excited over a press release saying that someday somehow they might make clean power cheaper than coal. I'm waiting for some real results.
- ntoshev 19y agoI just realized that if you are willing to mess with the system behind the scene, you can drive the community where you want. An easy systematic way to do it would be to make votes of people who vote like yourself (or a trusted group of people) automatically matter more (think kNN). In this way they are likely to push stories you would like but haven't noticed yet. I am sure this would be perceived as unethical, but I have no idea where the line between such an approach and normal editing/moderation can be drawn. Of course you can make personalized rating for everyone based on such a recommendation system. But you can also make your taste the default (when someone has not rated enough items) and in this way work to avoid fragmentation in the community.
- lucraft 19y agoThis comment reminds me of the time there were two defence related stories in The Daily Telegraph. Story #1: in approx 10 years time the Royal Navy will have no aircraft carriers, due to procurement failures, and for the first time since WW2 the UK will be unable to project air power around the globe. Story #2: British Army soldiers have been issued special underpants for war zones. They ran in the same week. Which do you think was put at the front of the newspaper with a large billing, and which was slipped in as a note on page 25? You get one guess.