OK, jlmale0, are you working on requirements or marketing for a product in this space? You can tell us, it's OK.

If you get a developer's license, you have to sign the confidentiality agreement from hell, and you can't even tell anyone what you signed. Apple gets 30% off the top, they get complete veto power over your apps, and they aren't just restricting their review to malware. If your app competes with something they want to do, they can just block it. There is no appeal. With phones, maybe you can justify the lockdown. And the economy sucks so bad that many developers will figure that indentured servitude is OK.

How is that relevant? Those dozens of users can use the proprietary Linux nVidia driver, which is not being discontinued, and has a lot of the same code in it as the Windows driver. The discontinued driver is an obsolete 2-D only hack; free software purists can use Nouveau, so who cares?

By this point, Nouveau beats the old nVidia open-source driver, so everyone would want to run either Nouveau or the proprietary nVidia driver. There's no real reason to support the obsolete, limited xf86-video-nv any more (though it's not going away).

If they do it right, it will fix the problem, because there's a sensor that detects pressure. If they don't do it right, it won't.

If you don't get the update, and your car electronics surge your speed to 100 mph but you refused the update that makes the brake pedal unconditionally override the acceleration, your heirs can blame the software, because you won't be around to do so. Take the freaking update. Oh, and the problem wasn't the floor mats.

Yes, you can buy a developer's license and hack away. But if you want to share your work with others, you are forced to go through an arbitrary review process before you can distribute it (the "app store"). Apple actually forbade people to distribute an e-book reader app that would download public-domain books from the Gutenburg Project because the Kama Sutra wasn't blocked.

Certainly there's a role for appliances, but the iPad appears to be aimed to compete with netbooks. I'd rather see completely programmable platforms with elegantly designed hardware and multitouch capability, and without a central authority approving of each program (or "app") before it can be distributed.

Don't give your kids Tcl; of the scripting languages, Python will be a lot easier, and the fact that it has "advanced concepts" is a plus, not a minus, as they don't get in the way and solve problems that the programmer would otherwise have to deal with. Basic is good if you want to teach kids to write rats' nests of GOTO statements.

"Even though the 1 TeV barrier per beam was first broken a week ago ..."

Um, no. There's nothing magic about 1 TeV. It's not a barrier.

Mach 1 was a barrier, because the aerodynamics is very different for a plane flying faster than the speed of sound. This means that new design principles had to be worked out. But nothing magic happens when you ramp up from 0.999 TeV to 1 TeV other than the flying champagne corks.

Likewise, new principles (optical proximity correction and phase shift masking) had to be invented so that we can manufacture ICs whose feature size is smaller than the wavelength of light (UV actually) used to expose the masks. That's an example of a barrier being broken.

But Slashdot should disallow the use of the word "barrier" just because a round performance number has been bettered. Alternatively, we can all just mock the editors every time they do it; you decide.

She spent two days in jail over this nonsense, and she talks about how nice they were? What a good little citizen. Perhaps she'll turn in a few friends to get in better with the authorities.

The music is public domain (because it's from an older song). The lyrics are copyrighted. So you can legally hum it in public for commercial gain.

I review papers for technical conferences. I regularly try to keep papers out of the publications. It's a necessary part of the job, because the acceptance rate is typically 25%, and because most of the papers are junk. Scientific publications are not free speech platforms; to be published, an article has to meet the standards and it has to advance the state of the art of the field.

The bar for skeptics is always going to be higher. Otherwise we'd have to rewrite the chemistry textbooks every time some student messes up his lab assignment, because this will produce data that contradicts the theory.

GPLv3 is actually more forgiving than GPLv2 of accidental violations. GPLv2 says that you forfeit the license and you need the copyright holder to reinstate it. GPLv3 provides a mechanism to correct the violation and have your permissions automatically reinstated. If you're producing a product that does DRM, you'll need to avoid GPLv3.

Under GPLv2, if you violate the terms, you lose the license, meaning that you can no longer copy or modify the work at all, and there is nothing in the GPL (v2) itself to get the license back. However, the copyright holder can forgive the violation and reinstate the license.

Likewise, under GPLv3 the copyright holder can give you additional shots to get the license back.

It's important to remember that the copyright holder's powers go beyond the terms. This does create problems for projects with hundreds of copyright owners, like Linux: if you violate the copyright, you apparently need the forgiveness of every Linux copyright owner, or, in the case of the dead contributors, their heirs, or you can never distribute a Linux kernel again (I suppose you could try to make a cut-down kernel without the contributions of the more unforgiving developers). Some might see this as a feature rather than a bug, though.

GPL code can import BSD code; it's only improper if the copyright notices are stripped off (which has happened, so you're right about that).