Only a fool "trusts the Internet" - especially Wikipedia.

It's funny, the other day, I was hanging out with a group that included several pretty top-level IT and networking folks, including some leading CS academics. Not one of us uses internet banking, or allows access of any kind to any of our financial accounts over the net. On the rare occasions that companies force the use of the Internet, the general response is to enable access only long enough to do the job, then destroy the Inet access account (best), disable net access (2nd best), or set the password to random gibberish that even we don't know or keep a record of. This forces a long, manual process to "reenable" the acccount that cannot as easily be done by an impostor. None of us "trust" the Internet, I guess.

That was a real eye-opener for some of the younger "Internet-savvy" group, who all of a sudden realized that maybe they were opening themselves up far more than they realized, especially in a world where every WiFi network, even with WPA2, is now as open as the one at Starbucks...

Actually, there's probably more money and effort focused on trying to build grid storage than there ought to be, given that there's really no technology known that's capable of doing the job in a generally viable way. There's a name for that: WOMBAT - Waste Of Money, Brains, And Time.

(Not saying we shouldn't be looking at all, but realistically, grid-scale storage requires technologies we simply don't have, and largely, can't yet even envision or propose. We're a smart society with a few centuries of intense technology and engineering development under our belts, and there is no known viable solution to this problem. If there was, then billions, or even trillions, of dollars would be flowing into it. This isn't like most hard problems, which can be solved by throwing enough effort and money at them - we really just don't know how to do this!

For all its faults, Hydrogen may be the best of the bad options - but the most (only?) economically viable source of hydrogen at large scales today is natural gas. Both environmentally and from an energy loss point of view, you're better off just burning the natural gas (our cleanest fuel in the first place) than taking the hit converting it to H2. Any effort to split water will result in H2 that is *much* more expensive than making it from natural gas, especially given the benefits of the fracking revolution - water is an *extremely* stable molecule...

Yeah, sounds simple, right? Just store it! First, even the best solar systems today are not economically viable without huge government subsidies unless you live on an island and have to ship your fuel in, so really, you're upside down before you've spent a dime on storage.

Secondly, with any known and viable technology storage is *really*, *REALLY* expensive on a grid scale. For all practical purposes, it's fair to say that there is NO known way to do it in most locations. (The dangers of gas-pressurized reservoirs may well be orders of magnitude higher than fracking at its worst, and very few places have geography that allow pumped hydro to be even marginally cost-effective.) Batteries, supercaps, and the like still need another couple of orders of magnitude price/performance improvement to be viable.

Do the math, and you'll see that storage isn't even an option - the solar plant is barely viable even with subsidies (here in Texas, with cheap and readily available natural gas, solar costs 4-5 times as much per KWH, according to EIA's LCOE figures). Add in any kind of grid-scale storage at all, and the costs soar through the stratosphere, especially since most storage technologies have relatively short economic lives.

So yes, paying someone to take the power is actually the cheapest thing to do - not only in Germany, but many nights here in the US with wind power, too. There's just more capacity than demand, and since it costs the power companies to deal with that, they justifiably want to get paid to offset the costs and inefficiencies of having to shut down and spin up their conventional plants.

I don't have any mod points, and have posted on this thread anyway, but LISTEN TO THIS GUY (brambus).

Unlike most of the armchair experts here, brambus is explaining *exactly* why the German grid is broken and why it will eventually fail - at this point, I think the only questions are "When?", and, "How bad?" The tariffs that led to all this investment in solar et al are completely unsustainable over the long haul, and everyone has known that all along, but like the actual climate record, it didn't fit with the narrative and had to be ignored.

If Germany is really lucky, they'll get by with some scary but not-totally-grid-meltdown failures that might finally kick some sense into the Greens and others who think they can legislate reality based on wishcraft...

It's sad, but they're going to have to learn what I taught my kids: You are free to choose your actions, but you are NOT free to choose the consequences of your actions...

And 30 years is best-case. In the real world, the output of quality solar panels at around 25 years will only be about 20% of their nameplate rating. That last 5 years is really just trying to eke out enough additional energy production to get positive over the entire life of the array.

Although tight, the economics are workable with good quality panels. Unfortunately, the crappy Chinese panels that now dominate the market are starting to show significant failures (backing delamination, which results in water ingress, destruction of the panel, and leaching of heavy metals into the environment) even BEFORE 10 YEARS. If that happens, you will NEVER, EVER break even on your solar plant.

Nobody mandated solar, people just decided it would work and be profitable. Germany got a lot of wind power built as well, but apparently solar also works well enough to be worth the investment.

Germany's problems are entirely of its own making - the government wrote laws that required the power companies to pay solar generators at rates that are often over 3X the going rate for electricity. Not surprisingly, a LOT of people took them up on that deal. This works sort-of-OK until a big squall line blows over and you lose a hundred megawatts in a few minutes (it's worse than that really, since sites that were exporting power to the grid now need to become consumers, so demand increases simultaneously even faster than the loss of supply!)

Germany is now the global poster child for grid instability, and I suspect they'll get bitten hard before too long - you can't keep up that balancing act forever, especially with declining spinning reserves, and no incentives for power companies to keep them at the ready. In the very near future, if Russia pulls the plug on natural gas at the same time as a major storm front, all of Germany will go black...

Not really. I've been working in the solar industry the last five or six years, and the short answer is that solar only makes sense without subsidies in places where you simply can't get energy from other sources - mostly islands or other areas where there are no fossil fuel resources nearby.

Again, that's not really true - depending on whose numbers you use, solar and other renewables are subsidized at a rate that is at least 25 to 50 TIMES that of any other energy source (including nuclear) on a per unit energy basis (which is really the only sensible way to even attempt a subsidy comparison.)

From a WSJ editorial based on the US govt's EIA figures (http://online.wsj.com/news/articles/SB10001424127887324432404579051123500813210):

The U.S. Energy Information Administration estimated in 2010 that fossil-fuel subsidies amounted to $4 billion a year. ... Renewable sources received more than triple that figure, roughly $14 billion. That doesn’t include $2.5 billion for nuclear energy.

Actual spending skews even more toward green energy than it seems. Since wind turbines and other renewable sources produce much less energy than fossil fuels, the U.S. is paying more for less. Coal-powered electricity is subsidized at about 5% of one cent for every kilowatt-hour produced, while wind power gets about a nickel per kwh. For solar power, it costs the taxpayer 77 cents per kwh. (Emphasis mine)

Don't get me wrong - I'm not opposed to solar, in fact, I favor it - but the fact is that solar and other renewables are not economically viable without subsidies. This is why the Original article is important - Germany has subsidized solar to the point that it's now a sizable portion of the German power grid. Unfortunately, renewables are NOT a replacement for power plants, since they literally only work at the whims of the weather. That means you still have to keep enough power plants in operation to meet peak demand. Therefore, letting the gluts determine prices is folly (this is why West Texas wind energy actually often has a *negative* price - you literally have to pay the grid to take it at night.)

And this distinguishes it from the US how?

Here in Austin, bicycle travel is subsidized to ridiculous degrees - new bicycle lanes are reducing 4-lane roads to two-lane all over town in a blatant and brazen attempt to botch traffic so badly that voters will finally approve the light rail boondoggle the city council has been drooling over for decades.

The ONLY thing mass transit does well is offer exceptional opportunities for graft, cronyism, and corruption.

Sorry, they are rarely if ever a public good - building roads is almost always far better and more efficient than another multibillion dollar mass transit boondoggle.

The only thing mass transit (especially "urban rail") excels at is creating the perfect environment for increasingly large-scale graft and corruption.

Although there may be one somewhere, I'm not aware of any mass transit system that breaks even on fares, nor am I aware of one whose cost even had the same number of zeroes promised when hoodwinking the voters into paying for it - Cost overruns of 10-100X are *ordinary* for mass transit projects, with powered train cars now costing more than F-16s!

Since mass transit systems never wind up being able to pay for themselves, mass transit is really just a taxpayer-funded subsidy for those who benefit from ridiculously dense, family-hostile, and outrageously expensive real estate and development.

Mass transit is one of the very best reasons to hate what most large cities have become. The cities without much in the way of mass transit are inevitably safer and more livable, and have much more positive and healthier mental attitudes. And yes, this is one of the few things that can destroy the dynamism even of cities like Austin...

Why is is that every successive generation is forced to re-learn the truth about the way people work? Even before I had any gray hair, I always sought to work on teams with widely mixed ages - as a young guy who thought he knew everything, I at least had a *chance* to short cut some painful and expensive lessons because the older, more experienced guys were usually more than happy to share their wisdom.

As was pointed out above, there is truly nothing new under the sun. (Seriously - even if you have zero inclination towards Judaism or Christanity, you really must read the Book of Proverbs sometime, just to be culturally literate and understand where so many of these phrases and sayings actually came from - it's a book of 31 short chapters, so one a day will knock it out in a month. My guess is you'll want to start over then, but YMMV...)

Today I intentionally build teams with a mix of ages, as it's by far the most important kind of diversity. (Yes, I know that's a very non-PC thing to say. Get over it.)

BTW, if you're looking for a more modern writing dealing with the issue, there is no better book than Fred Brooks' The Mythical Man-Month, which should be required reading for *anyone* working on, with, or around any software project, anywhere, ever.

Russian stuff varied a lot in practice, but some of it was actually brilliantly simple. Like Rutan's brilliant simplification of landing gear for a flying spaceship, some Russian designs wind up being better sinply becasue they *avoid* the problem rather than trying to solve it head-on.

Two examples I saw when working at NASA JSC in the 90's:

1) The Russians used a simple low pressure cooling system for their space stations and the original design of their ISS modules. This allowed them to easily use freeze-proof, but toxic coolants - since they operated below cabin pressure, if there was a leak, it was air into the coolant, not the other way around. The US approach used pressurized coolant, was insanely complicated by comparison, and *still* had the potential to freeze solid.

2) US spacesuit gloves are ridiculously complicated structures with many layers and exotic materials and parts laboriously assembled to make sure that they won't leak, or if they do, they'll self-seal, etc. so that a glove rip (a likely point of damage) won't lead to loss of suit pressure. Maintaining any dexterity in the glove while doing this is an obvious challenge with the many redundant layers. The Russians, on the other hand, use something more like a thick rubber glove (modified to avoid inflation effects), and a simple inflatable cuff that seals off around the astronaut's wrist in case of a leak - turns out that a full vacuum in the glove will blow a bunch of capillaries in your hand, turning it red for a week, but you'll be fine the week after. This gives a glove with more dexterity, at a cost that's only a tiny fraction of a US spacesuit glove...

I can't think of a single good technology that originated at Sun

ZFS, dtrace ?

Yep, those two, and how about network file sharing in general, including the various versions of the NFS protocol, and the YP/NIS/NIS+ systems that provided (for their day, anyway) secure and scalable directory services and access control?

How about bringing the best of both the Berkeley and AT&T System V Unix worlds together, with the guidance of the author of BSD, one of the most brilliant computer guys ever?

How about the first 64-bit hardware and OS that didn't require you to completely rewrite your apps and libraries to take advantage of that great new hardware? How about compilers from a computer manufacturer that actually didn't completely suck?

How about the open source graphical user environment (OpenWindows) that in the early days of Linux, finally gave it a GUI that didn't suck, and arguably transformed it from a schoolboy's neat hack into an alternative OS that could grow to run a fair fraction of the world's computers?

How about the very concept of corporate-sponsored, open source software in the first place? (Not just trivial fluffy stuff, but the actual guts of the system and services that run the computer and the network, including several mentioned above, eventually extending to things like OpenSolaris, Spring, and OpenStep.)

How about supporting networking and networked apps from the very beginning? - How about realizing that "The network *IS* the computer"? How about being one of the very first to adopt and support new high-performance networks? (3Mbps ->10Mbps Ethernet, FDDI, built-in 100 Mbps Ethernet, ATM, FibreChannel, etc.)

How about the first reasonably priced, small, scalable, lighting-fast storage array, with hot-swappable disks and compatible with the expensive industry-standard volume and disk management software, but also usable with free/inexpensive Sun "Solstice" alternatives?

And then, of course, there's Java itself. I'm not a huge Java fan these days, given the rise of things trying to do its job better, but there's no question Java is solid and runs important chunks of the world, and is largely responsible for the widespread adoption of object-oriented programming, which was previously a niche thing for the Smalltalk and LISP guys.

Oh, I almost forgot...

How about mice that required you to not twist their mirrored mousepads? (yeah, I know, SGI used those, too...)
How about the ridiculously goofy and expensive 13W3 video connectors and cables?
How about the rackmount Ultra servers that destroyed the CD tray if you ejected a CD with the door closed?
How about computers that really *were* the size and shape of a Pizza Box, but were strong enough support an 80-lb 24-in HDTV monitor on top?

Like they say about people telling Woodstock stories, you obviously weren't there...

I was at Sun doing market development in the healthcare and petroleum "verticals" when Java came out. I'm telling you, the interest was staggering. I once spoke on Java at a local JUNIOR college weekend CS/Internet interest forum to nearly a THOUSAND people, including top IT staff from NASA and all the major oil companies. I just broke the awesome oceangoing coffee mug they gave me about a year ago.

I can tell you that although we all realized Java was a good implementation of some great ideas, we were pretty much taken aback with the Java frenzy that ensued, and quickly moved to leverage it for all it was worth. (With a couple of decades in the rearview mirror, it's easy to forget how revolutionary Java really was at the time, and how hungry the world was for what it offered - namely the most open cross-platform platform and programming environment anyone had ever seen. It didn't hurt that the Java wave lined up really nicely with the 64-bit UltraSPARC architecture's amazing price/performance.)

It worked - Java was HUGE for both reestablishing Sun as a power player in technical and scientific computing, but also breaking into other lucrative markets we'd been frozen out of, including finance and healthcare - Before Java, Data General had far better name recognition than Sun - I literally met with a BIG heatlhcare CTO who's first question was, "So you're with Sun OIL?" He didn't even know there was a computer company called Sun. Shortly after, he was leading a transition to Sun hardware and software across his entire company. We got him hooked up with the right talent to integrate several critical Java-based products and he saved millions the first year, even after all the switching costs.

Trust me, we could have sold Java seats, no problem, although being free certainly helped its popularity and stood out from other "enterprise-capable" languages. The big mistake was when the programmers took over and turned a great system focused on cross-platform results and networked computing into something that tried in vain to check every box on the academic CS geeks' wet dream list, and the simple but vital stuff (like say, nailing down a single place where one could expect to FIND a JVM/JRE of a particular version on any given OS platform, to name one example of thousands) fell by the wayside.

Sadly, Java's never really recovered from the bloat it acquired in trying to be everything to everyone, but it did blaze the pathway for others, including what we called "Java with semicolons": the JavaScript that rules the web now and for the foreseeable future...

Sun was an amazing company with amazing people doing amazingly innovative things (NFS, YP/NIS/NIS+, Java, same binary desktop-to-supercomputer with transparent 64-bit support (compare Sun's 64-bit transition to IBM/HP/DEC's 64-bit cluster foxtrots - Sun's thinking here continues to fuel the current ARM revolution). There were some stinkers, but overall , we'd all be better off with Sun's innovation still pushing things forward. In a lot of ways, Sun was a better Apple than Apple when it came to "doing it right", especially back in the Java days, when we passed on actually buying Apple...

This is the really scary aspect of this decision, and it seems to be lost on most of the commenters here -

In order for Google (or any other provider of information services) to even putatively be *able* to delete all information they have regarding you, they first have to have all information about and relating to you tagged with your identity.

This effectively means they must (by law) keep an auditable log of everything you ever do on the net.

What could possibly go wrong with that? I'm telling you, Hitler and Stalin only dreamed of the power that big governments and big corporations will have to monitor, control, and even "disappear" us in the very near future...