Tiles unlike icons are supposed to be

a) live so they are presenting useful summary information
b) variable in size, so the amount can vary

The legacy stuff doesn't make use of the Window8 and Microsoft unfortunately itself doesn't install; by default some of the really cool applications that use Windows 8; like their Bing based applications.

Look at GP. Those are small amounts. See what happens if some retailer were doing 10k $100 cards / day per location paid for in cash.

I understand. But they are analogous. In a Hypervisor the real kernel doing most of the work is out in userland while there is a tiny "kernel" which runs part. Moreover the device drivers are split.

Ap X uses device Y
there is no device driver involved X uses a virtual device driver. The virtual device driver then talks to a real device. Which is pretty close to the Microkernel design.

I'd say that HURD is just a failed product and not read too much into it. QNX has been out for a long time with a terrific microkernel that offers real advantages. Its about to become too dated as their isn't going to be the funding to make the move to 64 bit but... Moreover arguably the virtualization OSes are essentially microkernels where a virtualization system acts as a mini kernel, multiple non-monolythic kernels operate on top of them and then branch out. Virtualization hasn't exactly been a failure. There are substantial advantages to both approaches and frankly Linus ended up introducing a modules system to capture many of the advantages of microkernels himself.

If they don't record the information though it is called money laundering. American Express got nailed with that in the 1990s for Traveler's Cheques they were letting people buy in the USA, lose in the USA, and then getting them recovered in a South American American Express location with no record of the who.

It means high prices (have you seen the prices on Windows Server and/or Microsoft Azure lately?

Yes pretty reasonable compared to Oracle, IBM and similar offerings. Higher than open source alternatives.

It means ignoring customers and forcing bad implementations on them (*cough*Metro*cough*) and then taking forever to admit it was a mistake and fix it (when is Windows 9 due out? Next year sometime?).

IMHO the mistake was not forcing it more by making touch and/or digitizer tablet mandatory for Windows 8. The problem with Windows 8 is that people insist on running on Windows 7 hardware.

Huh? The Power8 are 22nm same as Intel Haswell and frankly in most ways better than the Haswell.

I absolutely agree with you. I have some doubt though that SICP needs to be a first course.

Embedded, ROMs, low level OS components... is obviously the old paradigm. Most of an OS or VM though I'd say is going to be high level library manipulation. As for writing a library, I think it depends on the library. Most libraries today are highly dependent on other libraries.

Scheme was developed from LISP to prove the possibility of constructing a language from the ground up using a Meta-circular evaluator. That was important for SICP because it meant that every student knew how to create a language using arbitrary primitives, a DSL. There is no way in an 1 year introductory Python class the students would be anywhere near ready to implement a Python with different primitives. You can argue that the entire LISP philosophy is the wrong approach to solving complex problems, and I think history has perhaps show that, but arguing it was just stupid is a bit much.

As for going 18 levels deep that probably should be broken out about every 5 levels into smaller simpler functions.

As for arrays:

    (array dimensions element0 ...) -> array
    (array '(2 3) 'a 'b 'c 'd 'e 'f) -> {Array 2 3}
That doesn't seem that hard.

as a functional language, new students must first learn to think along a different paradigm, one opposed to how they had been thinking their whole life, before they can begin to comprehend the basic concepts they're supposed to be learning

Most students can handle Excel, a functional language. I think breaking them of their bad habits is a good thing for an intro course. It puts most of the students on the same level regardless of background. In a week they won't get there, in a year certainly.

Mostly if I were teaching an intro course today I'd probably go with Python. But I don't think it is nearly as clear as you do. Haskell for example (which keeps most of the essence of LISP with giving them more modern concepts) would be a consideration.

The people who designed the SICP curriculum felt it wasn't teaching the right paradigms. SICP was built around a world where a programmer wrote small programs and tied them together. An individual programmer could really understand an entire production program. Today's programming world involves programmers using massive and complex specialized libraries with far more large group projects. SICP/Scheme didn't train people for that sort of environment. They needed to switch from "what data-structure would best accomplish this goal" to "which library would best accomplish this goal" and Scheme encourages much the opposite.

SICP is probably still the best programming concepts book ever written but those concepts are less important than they used to be.

Python doesn't really support functional programming. A good multi-paradigm educational language which also has a great textbook is Oz.

What are you talking about? Interpreted languages let you reuse data structures for performance. What he did was a performance hack. If anything in languages where getting performance is more difficult we'd expect more, more complex and more subtle bugs like the SSL one.

There is no way you are going to have a legal framework where knowingly distributing materials for which you do not have a license is going to be no big deal. There is no way you are going to have a legal framework where judges are going to be the first step in any copyright dispute.

It is not going to happen.

My point wasn't that it was likely that anybody filing a counter-claim would be sued. I merely am saying that it is possible, which means they're taking a risk, and they probably stand to gain nothing personally by taking it. That is likely to have a chilling effect.

No question it has a chilling effect. Right now we have a very wide open casual publishing culture and a legal framework designed for a formal publishing culture. At their point of contact things are messy. My point is that the DMCA isn't the problem. The problem is that if people are worried about getting sued for content they need to respond to cease and desist.

If we did what most of the other people recommend Qualcomm would be suing that out of the blue with no warning.

A cease and desist order on other topics can turn an unintentional act int an intentional one. DMCA and safe harbor is just a specific example of that. Cease and desist create intent, that is their point.

The easy way to see that is if the website presented information to the ISP of their licenses for everything in advance the DMCA wouldn't require any action other than passing that information back. It is legal to distribute content assuming everyone is properly licensed without establishing that first. That's very generous. Of course it is not legal to distribute content once someone contests the content and the originator won't at least claim they are licensed.