+1000

I think this shows that Google decompiled Sun's binaries. If you run a typical Java decompiler, you lose generics, local variable names; some initialization gets moved around; and certain things get inlined. Looks to me like there's evidence of all of the above coming across from Sun's source to Android.

Now, decompilation is a murky legal area I think. I guess it falls under the rubric of reverse-engineering, but I have no idea how that law handles such practices in this context. I suppose Oracle are operating with the knowledge that Google likely decompiled class files and have chosen to pursue the claimed that they copied the source. Is it because decompilation is harder to prove? Does it carry less legal weight than a straight copy or a near-copy? Hm...

That's what lightweight widgets are (a la Swing) -- control presentation across platforms and don't rely on native libraries and threads for creating and destroying native UI elements (peers). This is not a new idea. One of the first implementations of this for performance reasons that I'm aware of was in Swing's grid widget (JTable), where a single lightweight component is reused for drawing each cell that uses the same rendering logic. This is in contrast to how it might have been done using a native widget, which would have to be instantiated for each visible cell separately.

There's nothing magical about native widgets -- at some point it all comes down to drawing stuff on the screen. It just depends on which library does it and how efficiently (threads, memory, hardware acceleration, etc.).

I don't understand why they wired regenerative braking on the front wheel to the rear wheel brake lever. I understand that it lets them have a single brake lever for the whole bike, but generally speaking, most bicycles and motorcycles have separate controls for each wheel's braking for a good reason.

This is quite interesting for me coming across this discussion as just yesterday I happened to be talking to a K-2 educator, who has developed an introductory programming curriculum based on Alice and Scratch environments. I personally have no first-hand experience teaching kids, but was fortunate enough to be exposed to a variety materials and teachers early on. I'm still not sure if they fostered my interest or gave an outlet to something I had, or perhaps a combination of both.

I distinctly remember one of the first things that captivated me was a translated Brazilian comic book that accurately and progressively portrayed the basics of AI and robotics. It was intelligent, humorous and seemed to be just at the right level, where I was neither bored nor overwhelmed. I read and reread it countless times, always wanting more when I got to the end. Other important moments seemed to be tied to more conventional classroom settings using programmable calculators, BASIC, Pascal and the like. Looking back, it seems not having a computer at home with very limited computer lab time, it required an intimate understanding what your program did and debug it as much as possible in your head -- very similar to the world of punch cards. I don't remember now under what circumstances I acquired a C++ book, but I remember trying to understand the concepts, reading it alone, getting frustrated, putting it away, only to come back to it months later and trying again. I think it was a year of doing it before I had a huge "Aha" moment, when I finally "got" objects. The irony was that I didn't actually get to program in C or C++ all that much later in life, but that incredibly dry and difficult book with virtually no access to a computer got me to the point, where I didn't look back and knew for a fact that I enjoyed learning about programming. What's also interesting is that while I also liked natural sciences and mathematics and did okay under lots of positive pressure and great teachers, I would get frustrated and walk away much more easily and it was that pressure that kept me in it 'til the end, whereas with programming, I couldn't get enough.

I have no idea if this kind of path is right for anybody else, or if it can even be reproduced nowadays. But I've always had a feeling that there's something to teaching programming without computers themselves having primary focus. The teacher I mentioned told me that while he himself does use a computer lab for his curriculum, there are experimental classes, where kids are being asked to do various puzzles and problems designed to steer them in the direction of thinking in terms of algorithms. He also said that it's much more about the teaching method than the tool. Understanding age-appropriateness and keeping one's finger on the pulse of their motivation is key -- seems obvious, but not so easy to do well in practice.

In 2009, Yamaha has come out with what I believe is the first production cross-plane I4 engine for their YZF-R1 motorcycle. Previously, they were using it in prototype racing (MotoGP) in YZR-M1. The reasons for using a cross-plane crank in an I4 engine are numerous, mostly specific to motorcycles.

I seem to recall this being brought up by a military commander, whose name escapes me at the moment, right at the onset of the current war in Iraq. Since then, according to this article, they are just now in the process of implementing encryption and it sounds like they're not planning on doing it for the older units at all. When I first read about this, the biggest worry was that not only the video, but the control of the drones could be intercepted. I really wonder.

Note, that in the US, debit and credit card transactions are expensive for the merchants and credit card transactions are not at all guaranteed due to possibility of charge-back. Businesses like them for large transactions, where many forms of identification and authorization are used and there's little chance of charge-back or fraud, but with medium or small scale transactions, cash is king. Of course, it's not always possible, but whenever it is, you can almost always get a "cash discount" from a small business because, if for no other reason than because they can hide it from their books.

Anyone else find that "Arty" MS mouse should really go by the name Mickey mouse?

A 10% increase in top speed would require roughly a 21% increase in power and/or efficiency, provided aerodynamics were the same. Granted, these guys probably worked on the aero qualities of the car, but still, let's not call it only 10%.

Doesn't need to be this way. If the automated system works as advertised, then it can be thought of as a simple cross-compiler and the code base can continue to be maintained in COBOL. I can see how this would make sense if they wanted to save money on the hardware.

I worked on a similar project for the mental health industry in the 1990's. Same exact problem. We'd get 3 or 4 respected psychiatrists in the room and they'd butt heads for an hour on how to represent in the system seemingly the most basic of concepts. I don't know what the answer is, but it seems like we need to be very watchful of how flexible the solutions we come up with are and at the same time not let the medical professionals off the hook completely when it comes to reasonable standardization. There are examples already in other countries that have waited a while and recently implemented EMR systems. Surely, they're worth studying.

Judgmental much?

What are your criteria for which insurance is better? If you believe private insurance companies will post a higher profit and will provide more timely care at a premium for those who can afford it, I'd probably agree with you. But as far as decent care across the board, not downgrading preventative care, not incentivizing doctors and hospitals to perform as much tests as possible, not driving GP's into a an increasingly more difficult business, I'd say a big public plan would be better.

All that means is that either SS tax needs to go up or retirement age needs to go up, which sucks, but makes sense, given our current demographics shift. It doesn't mean the entire concept is unsustainable or deeply flawed.