Microsoft never claimed that their offering was certified. Their claim was that Google was lying by claiming a certification that Google didn't have.
Apparently some people who have more hatred for MS than reading comprehension skill have twisted this into a claim that Google was pretending to have a certification that MS already has. That's not the case.
Actually, I don't recall a single place where MS said their offering was FISMA certified. They weren't saying "Our offering is and Google's isn't, so choose us!" they were saying "Google is saying their oiffering is certified but it's not; they're lying to you." So far as I've seen, this is true. Microsoft never tried to hide that their offering wasn't certified yet, they're just a vendor calling out their competitor for lying to the client (the government).
Exactly. The process of evolution is gradual. Speciation doesn't occur in a single generation, or even in a single lifetime.
Consider equines. Horses and donkeys clearly share an evolutionary ancestor. In fact, they haven't even completely diverged from that ancestor; despite "obviously" being different species, they are inter-fertile. The offspring (mules) are infertile, so it is reasonable to call horses and donkeys different species; they can produce live offspring, but those offspring are a genetic dead end.
OK, how does that relate to my point? Well, sometime many millennia ago, there were a group of equines that, although not exactly like today's donkeys, were close enough that you would call them donkeys. There was a similar group of "horses". Here's the weird thing: they were the same species (interfertile and producing viable offspring). Somewhere over the millennia since then, the two groups, breeding primarily within their own group and not between groups, reinforced certain traits to the point where cross-group offspring were no longer fertile.
The question for you: how the heck do you define where speciation occurred? Was it when the (still interfertile) groups started moving apart? Was it the first member of each group that could not produce fertile offspring with more than half the potential mates in the other group? Was it when there was one member of each group which were mutually incapable of producing fertile offspring with any descendents of the other? For that matter, how do you define thr groups themselves? There were probably some fertile proto-mules for a while, which didn't fit cleanly into either group. They either died out without reproducing or were merged back into one of the groups, the line would nonetheless have been somewhat blurry.
Now, next question: how do you determine, from the fossil record, where that speciation occurred? Which of a bunch of old horse/donkey-skeleton-like rocks (that's all fossils are) was once an animal that gave rise to modern horses which can't produce fertile offspring with modern donkeys? How do you distinguish, from the fossils, that it was X, and not the parents of X, or the children of X, or possibly the specific children of X by Y? How do you distinguish that it was X and not X's sibling that got a slightly different set of chromosomes and was no longer able to produce fertile offspring with his or her corresponding member of the other group, yet went on to breed successfully and pass those chromosomes onto the other members of the group?
Seriously, demanding to see "direct ancestors" in the fossil record is absolute stupidity. I'm no biologist (as I'm sure any biologist reading my post noted) but I understand enough basic genetics to know that even with genetic evidence it's non-trivial to trace direct ancestry, and without it the task is nearly hopeless. Combine that with the way that most individuals never get fossilized, much less last long enough after fossilization to be found today (never mind the many fossils that we don't have yet; new finds are still occurring). Given all that, it'd be a minor miracle to have gaps of only 1000 generations in a direct chain of ancestry. That's enough generations for some pretty significant changes, when you're looking for incremental differences between a horse and a donkey. 1000 generations ago, your ancestors were recognizably human, but they still looked different enough from you today that you wouldn't have been able to call them a "direct ancestor" or not from fossilized bones - and they were probably still close enough that you'd have been interfertile!
WinCE has run on ARM for a decade. It's not that MS hasn't had the ability to run on ARM, it's just that until recently ARM wasn't seen as a suitable chip for a desktop-style OS. CE powers lots of handheld or embedded devices, from smartphones to RFID scanners, and I suspect most of them run on ARM (CE runs on a few other architectures too, but ARM has dominated that market for a while).
Now that there's a demand for more "serious" operating systems on ARM, they're porting NT (a much more capable kernel than CE) over to it. Have ported, in fact, apparently. Interestingly, they've also apparently also already ported the Javascript JIT compiler; I'm pretty sure you can't get the performance shown in the video out of a 1 GHz chip interpreting JS.
Close. Server 2008 (6.0) and 2008 R2 (6.1) are also on ia64 (Itanium). They've announced that Itanium support is being dropped for future releases ("Windows 8", whatever its NT version number will be) though.
Umm... what version did you expect MS to release following 9 (which came out ~ 1 month ago)? In any case, 10 is nowhere near done - all you can download of it today is a rendering engine review, much like what you could get of IE9 a year ago - so I'd hardly call it "accelerated releases". If Firefox releases at the rate they say they will, I suspect they'll be at 10 by the time IE10 actually ships too... but where it will have been a lot of small changes over the months, IE10 will be a monolithic step up from 9.
Call that good or bad as you will. I'm in favor of getting newer features faster, but speaking both as a software developer who understands the cost of testing a release, and a mostly-former web developer who prefers to have fewer browsers I need to test against, I think the IE approach is possibly more realistic.In the meantime, the "platform preview" builds will give the opportunity to try out newer IE rendering and JS engines as they add features, fix bugs, and improve performance.
That's interesting. It also has absolutely nothing to do with the subject, except for the very last sentence you quoted. When something is certified, neither a subset nor a superset of that thing also receive certification. If Google did in fact claim that GAfG had FISMA certification before FISMA certification was granted to GAfG, then they were lying.
Mind you, it has no real compatibility with Windows user-mode code, much less drivers (despite what people seem to think, even Vista could run ~95% of XP drivers without a hitch). It is best run in a VM, where the "hardware" is simple and it's easy to mess with things without breaking everything or needing another machine on hand. Remember, it's a research project, not a production OS.
Midori... nobody outside of Microsoft seems to know anything about it, but most of the speculation could be succintly described as "stupid" because it just doesn't make sense to have on OS that is going to break backward compatibility so hard when that same compatibility is one of Microsoft's greatest assets. There's nearly 2 decades of work in the NT kernel, and despite what Slashdot likes to have to say about MS software in general and Windows in particular, most of the core of NT is very, very good. Throwing all that away for something that wouldn't be able to run the majority of Windows software would be silly.
Sorry, this doesn't make sense to me. The event horizon surrounds the singularity. It is defined as the distance where, once an object is less than that distance from the singularity, no information will ever breach the event horizon again. This applies even to photons; the reason a singularity is called a "black hole" is because once light crosses the event horizon it will never return, giving the effect or an object with absolutely no albedo.
In other words, once something passes the event horizon, you *cannot see it* because any light reflecting from it will also have passed the event horizon, and gravity will curve the light back toward the singluarity before the light can escape. The last glimpse you will have of the object is just before it crosses the event horizon. I'm not certain what will happen to light trying to travel directly away from the singularity from a point outside the event horizon - it might get distorted somehow, but I don't have the physics to be sure - but it would get back to you. Once it crosses the even horizon, though, it's gone forever even if it reflects off (or originates from) an object just inside the event horizon.
Meanwhile, there will be a period of time after crossing the event horizon but before the object is actually obliterated (although it will be "destroyed" in the classic sense by tidal forces much earlier, as this star was). As I understand it, an outside observer can (relatively) easily calculate the time it will take for the object to reach the singularity, assuming its the velocity crossing the event horizon and the mass of the singularity are known.Howeve,r for the object itself, time would subjectively be dialated to such an extent that the journey would seem, if not eternal, close to it (again, I suspect that I don't have the physics to answer this one correctly).
Your numbers are suspect but your reasoning is sound.
One good example is actually in web browsing, something that every single smartphone does. On the desktop, IE9 (and possibly other browsers, I don't know) does its JavaScript JIT compilation in parallel to its page rendering, taking advantage of the ubiquity of multi-core systems today. There's no reason a phone couldn't do the same. Visit a HTML5 web page with some intereactive animations. One thread runs the network loop, downloading stuff (and spawning additional download threads). As things are downloaded, they get passed to other threads; HTML and CSS go to a rendering thread, imagies, audio, and video go to a decoder/playback thread, and JavaScript goes to a JIT thread and/or/then an execution thread (compiled or interpreted - sometimes its faster to interpret a small operation at the begiining of a page than to hole up everything until it has been compiled). Throughout it all the UI thread remains unblocked. You can easily see this scaling well beyond four cores.
It's not just about the web browser, either. All major platforms have apps that are essentially special web pages (HTML/CSS/JS) designed to act like apps. In fact, I believe this is still the primary development method for WebOS, although it's debatable whether it qualifies as a major platform right now. Even on non-web-based apps, there's plenty of potential for parallelism in a phone app. Everything from AI in games to network code in a media streaming app can benefit from more hardware threads.
Your first two sentences are contradictory. It looks like you're trying to create an "unpatent" that is to patents as "copyleft" is to copyright. I like the idea. However, I don't think you understand what you're talking about in implementation. Public domain means you *can't* put any restrictions on it. You can only restrict something if you have any control (for property, typically ownership is required) over it. Public domain means relinquishing all control. The only restriction on public domain is that nobody can take something *out* of the public domain. That's it. Once released, anybody and everybody can use it for whatever purpose they want, produce derivative works, restrict those works however they please, and there's not a single thing you can do about it.
The reason for the minor rant above is because I see in your post the same kind of misunderstanding that leads to people thinking that open source licenses are the same as being in the public domain. Even things like the BSD or MIT licenses, much less the GPL and its bretheren, maintain more control than public domain.
For the record, there *is* a music subscription service analogous to (if not quite as nice as) Netflix. For $15/month (less for long-term subscriptions), you can get a Zune Pass which includes all-you-can-eat music downloads. They're DRMed, of course, but the bitrate is respectable. The problem is that they won't play on anything but a Microsoft platform.
You also get 10 credits a month for free MP3 downloads, 320kbps. These are DRM-free, play in anything, and are usable on some music where MS hasn't been able to negotiate for Zune Pass streaming/downloads. If you consider a tyical cost of $1/song, those 10 credits make up 2/3 the cost of the monthly subscription by themselves.
Mind you, I think that's a stupid law. The maximum damage allowed should perhaps be relative to the crime. For example, somebody pirating a texting app should pay for a bunch of texts, as opposed to somebody pirating a texting app that lets them see their surroundings instead broadcasting a beacon to someobdy who runs them over with a car. However, I think in principle the idea of prohibiting this kind of thing is wrong. Impose whatever penalty you want on somebody who harms an innocent through such a trap, of course. That alone will keep the more excessive responses at bay. I just don't understand why somebody willfully violating the law should receive equal protection under it, though. Again, that's not to say they should receive *no* protection, but if I want to dig 10' pit traps to deal wiith those pesky kids trespassing on my lawn, I think I should have that right so long as the limits of my lawn are clearly delineated, and the traps aren't likely to cause extreme harm.
Windows Phone 7 does not allow side-loading by default, but there are several ways to enable it anyhow. The most legit option is to get a marketplace developer account, which costs a bit of money but also gives the right to publish on the marketplace. At the other end, there's ChevronWP7, which is still available on phone hacker sites. In between there are things like using LG's built-in registry editor to tweak the "allow sideload" option.
There is, for example, a NES emulator which is not in the WP7 marketplace but can be downloaded online and sideloaded.
Why spend all that time configuring when a clean install (of newer Windows versions) is about as fast, gets rid of *all* the crap as opposed to just the stuff you find, and gives you the disk space back as well?
Fine, so they move the keylogger into the BIOS/EFI. If you can't trust the hardware manufacturer (and clearly you can't, in the case of Samsung), there's nothing to do but examine every single component with an electron microscope to ensure that the RAM controller isn't shunting off everything coming from the keyboard buffer into a stream to the network card. I could write a rootkit that, installed on a hard disk's firmware microcontroller, is completely undetectable from the outside without physically removing the disk platters. If it sent info you might be able to intercept that info, but if it instead just did something like passively wait for a given date and then crash unrecoverably, you could (for example) bring down entire datacenters simultaneously - and there'd be no way a priori to know it was coming unless you directly examined the firmware (and not just what the FW reported itself to be, but the actual bits in the EEPROM or flash or whatever).
There are even several legit places to get the install DVD image. It's the license key that MS is really concerned about, not the bits. You can also use somebody else's DVD just fine, even if it's for a different edition (so long as it's for the correct architecture).
Right, because everybody knows that 90% of a game's code is in its UI and input system. Things like the game engine, AI, logic controlling elements in the game, resources, and netcode are completely irrelevant, right?
To be fair, WP7 doesn't support much in the way of netcode right now, and it's certainly not trivial to shift UI paradigms. However, that doesn't mean that the ability to use XNA, and resuse a lot of code as a result, isn't still quite valuable.
1).NET is a framework, designed to provide a rich set of libraries that work with a number of front-end languages and produce programs that are portable across platforms, without even recompiling. The only sense in which it differs from the original goal of Java is that it emphasized cross-language more than cross-platform, although both frameworks support multiple front-end languages and run on multiple platforms today. The biggest implementation difference between the two, framework-wise, is that.NET allows lower-level access (direct "unsafe" pointers) than Java, and was designed from the start to avoid some of the pitfalls Java encountered around things like generics.
2).NET allows *far* more versatility than most frameworks. It's not quite as good for metaprogramming as C++, or for data/code equivalence as Lisp or even Javascript, but it can get there easier than you might think. In the meantime, it supports everything from pointer access to functional programming, and you can mix a project written in C# that uses unsafe blocks to talk directly to driver code with one written in F# that has no side effects at all. You can use LINQ to easily and efficiently work with data collections in all kinds of interesting ways, Reflection to pull apart compiled assemblies and work with their parts directly, Attributes to easily extend the way that languages work... Hell, I'd say somebody who knows *all* the ins-and-outs of.NET probably knows more of how to solve any arbitrary problem than somebody who fully knows Java, C++, and LISP, and it's not only because the first guy actually knows *more* languages. I'd give the second guy the edge if he also knows at least two assembly languages.
Can you even read more than three sentences at once? His (Her?) issue isn't that Apple controls what ehy allow to be sold in their store, it's that Apple doesn't allow anybody to set up a competing store. If I wanted strippers I wouldn't go to a grocery store, I'd go to a strip club. If somebody tried to say "No, you can *only* buy things that the grocery store sells" then damn straight I'd protest (boycotting being a common choice of protest).
When you enable BL, the OS will generate a recovery key (a whole bunch of random alphanumerics; I think it actually has more entropy than the crypto key) and force you to store it somehow. Options include printing and writing to a flashdrive or other external storage (on a domain, it can be backed up by the domain controller). This recovery key works for any situation where the "easy" unlocking methods that you're using (for example, a TPM + smart card) don't work. It can be used to unlock or fully decrypt the volume on another computer, in case something goes drastically wrong.
What version of IE are you using that doesn't break those apps, but IE9 does? IE9 offers compatibility modes for 7 and 8 (the versions that shipped with the only OS versions which can run 9) and a "Quirks" mode that works with nearly all legacy sites designed for IE6 or older. We have no lack of stuff on the Intranet that either use corporate ActiveX controls or are desktop apps which embed MSHTML, and I think there's only one of them (which I've never used) that doesn't work correctly with IE9. I've been using trouble-free for a few weeks (RC) at work.
T-Mobile? What are you talking about? T-Mobile US carries the HTC HD7 and Dell Venue Pro, both WP7 devices, and has since the US launch day. They don't sell the DVP in their own stores, true, but they certainly do sell the HD7.
They don't advertise as heavily as AT&T, but it's certainly not true that they're only releasing this year.
Same in IE9. Tools -> Safety -> ActiveX Filtering. Blocks everything, including Flash and Java, just fine. It can be toggled on and off for individual sites quite easily; the little circle-with-a-line-through-it icon in the location bar (between the Search drop-down and the Compatibility Mode toggle button) works as a button to control it, and turns blue when filtering anything.
Slashdot Mirror
Microsoft never claimed that their offering was certified. Their claim was that Google was lying by claiming a certification that Google didn't have.
Apparently some people who have more hatred for MS than reading comprehension skill have twisted this into a claim that Google was pretending to have a certification that MS already has. That's not the case.
Actually, I don't recall a single place where MS said their offering was FISMA certified. They weren't saying "Our offering is and Google's isn't, so choose us!" they were saying "Google is saying their oiffering is certified but it's not; they're lying to you." So far as I've seen, this is true. Microsoft never tried to hide that their offering wasn't certified yet, they're just a vendor calling out their competitor for lying to the client (the government).
Exactly. The process of evolution is gradual. Speciation doesn't occur in a single generation, or even in a single lifetime.
Consider equines. Horses and donkeys clearly share an evolutionary ancestor. In fact, they haven't even completely diverged from that ancestor; despite "obviously" being different species, they are inter-fertile. The offspring (mules) are infertile, so it is reasonable to call horses and donkeys different species; they can produce live offspring, but those offspring are a genetic dead end.
OK, how does that relate to my point? Well, sometime many millennia ago, there were a group of equines that, although not exactly like today's donkeys, were close enough that you would call them donkeys. There was a similar group of "horses". Here's the weird thing: they were the same species (interfertile and producing viable offspring). Somewhere over the millennia since then, the two groups, breeding primarily within their own group and not between groups, reinforced certain traits to the point where cross-group offspring were no longer fertile.
The question for you: how the heck do you define where speciation occurred? Was it when the (still interfertile) groups started moving apart? Was it the first member of each group that could not produce fertile offspring with more than half the potential mates in the other group? Was it when there was one member of each group which were mutually incapable of producing fertile offspring with any descendents of the other? For that matter, how do you define thr groups themselves? There were probably some fertile proto-mules for a while, which didn't fit cleanly into either group. They either died out without reproducing or were merged back into one of the groups, the line would nonetheless have been somewhat blurry.
Now, next question: how do you determine, from the fossil record, where that speciation occurred? Which of a bunch of old horse/donkey-skeleton-like rocks (that's all fossils are) was once an animal that gave rise to modern horses which can't produce fertile offspring with modern donkeys? How do you distinguish, from the fossils, that it was X, and not the parents of X, or the children of X, or possibly the specific children of X by Y? How do you distinguish that it was X and not X's sibling that got a slightly different set of chromosomes and was no longer able to produce fertile offspring with his or her corresponding member of the other group, yet went on to breed successfully and pass those chromosomes onto the other members of the group?
Seriously, demanding to see "direct ancestors" in the fossil record is absolute stupidity. I'm no biologist (as I'm sure any biologist reading my post noted) but I understand enough basic genetics to know that even with genetic evidence it's non-trivial to trace direct ancestry, and without it the task is nearly hopeless. Combine that with the way that most individuals never get fossilized, much less last long enough after fossilization to be found today (never mind the many fossils that we don't have yet; new finds are still occurring). Given all that, it'd be a minor miracle to have gaps of only 1000 generations in a direct chain of ancestry. That's enough generations for some pretty significant changes, when you're looking for incremental differences between a horse and a donkey. 1000 generations ago, your ancestors were recognizably human, but they still looked different enough from you today that you wouldn't have been able to call them a "direct ancestor" or not from fossilized bones - and they were probably still close enough that you'd have been interfertile!
WinCE has run on ARM for a decade. It's not that MS hasn't had the ability to run on ARM, it's just that until recently ARM wasn't seen as a suitable chip for a desktop-style OS. CE powers lots of handheld or embedded devices, from smartphones to RFID scanners, and I suspect most of them run on ARM (CE runs on a few other architectures too, but ARM has dominated that market for a while).
Now that there's a demand for more "serious" operating systems on ARM, they're porting NT (a much more capable kernel than CE) over to it. Have ported, in fact, apparently. Interestingly, they've also apparently also already ported the Javascript JIT compiler; I'm pretty sure you can't get the performance shown in the video out of a 1 GHz chip interpreting JS.
Close. Server 2008 (6.0) and 2008 R2 (6.1) are also on ia64 (Itanium). They've announced that Itanium support is being dropped for future releases ("Windows 8", whatever its NT version number will be) though.
Umm... what version did you expect MS to release following 9 (which came out ~ 1 month ago)? In any case, 10 is nowhere near done - all you can download of it today is a rendering engine review, much like what you could get of IE9 a year ago - so I'd hardly call it "accelerated releases". If Firefox releases at the rate they say they will, I suspect they'll be at 10 by the time IE10 actually ships too... but where it will have been a lot of small changes over the months, IE10 will be a monolithic step up from 9.
Call that good or bad as you will. I'm in favor of getting newer features faster, but speaking both as a software developer who understands the cost of testing a release, and a mostly-former web developer who prefers to have fewer browsers I need to test against, I think the IE approach is possibly more realistic.In the meantime, the "platform preview" builds will give the opportunity to try out newer IE rendering and JS engines as they add features, fix bugs, and improve performance.
That's interesting. It also has absolutely nothing to do with the subject, except for the very last sentence you quoted. When something is certified, neither a subset nor a superset of that thing also receive certification. If Google did in fact claim that GAfG had FISMA certification before FISMA certification was granted to GAfG, then they were lying.
Singularity is open-source, available for download today for free. http://singularity.codeplex.com/
Mind you, it has no real compatibility with Windows user-mode code, much less drivers (despite what people seem to think, even Vista could run ~95% of XP drivers without a hitch). It is best run in a VM, where the "hardware" is simple and it's easy to mess with things without breaking everything or needing another machine on hand. Remember, it's a research project, not a production OS.
Midori... nobody outside of Microsoft seems to know anything about it, but most of the speculation could be succintly described as "stupid" because it just doesn't make sense to have on OS that is going to break backward compatibility so hard when that same compatibility is one of Microsoft's greatest assets. There's nearly 2 decades of work in the NT kernel, and despite what Slashdot likes to have to say about MS software in general and Windows in particular, most of the core of NT is very, very good. Throwing all that away for something that wouldn't be able to run the majority of Windows software would be silly.
Sorry, this doesn't make sense to me. The event horizon surrounds the singularity. It is defined as the distance where, once an object is less than that distance from the singularity, no information will ever breach the event horizon again. This applies even to photons; the reason a singularity is called a "black hole" is because once light crosses the event horizon it will never return, giving the effect or an object with absolutely no albedo.
In other words, once something passes the event horizon, you *cannot see it* because any light reflecting from it will also have passed the event horizon, and gravity will curve the light back toward the singluarity before the light can escape. The last glimpse you will have of the object is just before it crosses the event horizon. I'm not certain what will happen to light trying to travel directly away from the singularity from a point outside the event horizon - it might get distorted somehow, but I don't have the physics to be sure - but it would get back to you. Once it crosses the even horizon, though, it's gone forever even if it reflects off (or originates from) an object just inside the event horizon.
Meanwhile, there will be a period of time after crossing the event horizon but before the object is actually obliterated (although it will be "destroyed" in the classic sense by tidal forces much earlier, as this star was). As I understand it, an outside observer can (relatively) easily calculate the time it will take for the object to reach the singularity, assuming its the velocity crossing the event horizon and the mass of the singularity are known.Howeve,r for the object itself, time would subjectively be dialated to such an extent that the journey would seem, if not eternal, close to it (again, I suspect that I don't have the physics to answer this one correctly).
Your numbers are suspect but your reasoning is sound.
One good example is actually in web browsing, something that every single smartphone does. On the desktop, IE9 (and possibly other browsers, I don't know) does its JavaScript JIT compilation in parallel to its page rendering, taking advantage of the ubiquity of multi-core systems today. There's no reason a phone couldn't do the same. Visit a HTML5 web page with some intereactive animations. One thread runs the network loop, downloading stuff (and spawning additional download threads). As things are downloaded, they get passed to other threads; HTML and CSS go to a rendering thread, imagies, audio, and video go to a decoder/playback thread, and JavaScript goes to a JIT thread and/or/then an execution thread (compiled or interpreted - sometimes its faster to interpret a small operation at the begiining of a page than to hole up everything until it has been compiled). Throughout it all the UI thread remains unblocked. You can easily see this scaling well beyond four cores.
It's not just about the web browser, either. All major platforms have apps that are essentially special web pages (HTML/CSS/JS) designed to act like apps. In fact, I believe this is still the primary development method for WebOS, although it's debatable whether it qualifies as a major platform right now. Even on non-web-based apps, there's plenty of potential for parallelism in a phone app. Everything from AI in games to network code in a media streaming app can benefit from more hardware threads.
Your first two sentences are contradictory. It looks like you're trying to create an "unpatent" that is to patents as "copyleft" is to copyright. I like the idea. However, I don't think you understand what you're talking about in implementation. Public domain means you *can't* put any restrictions on it. You can only restrict something if you have any control (for property, typically ownership is required) over it. Public domain means relinquishing all control. The only restriction on public domain is that nobody can take something *out* of the public domain. That's it. Once released, anybody and everybody can use it for whatever purpose they want, produce derivative works, restrict those works however they please, and there's not a single thing you can do about it.
The reason for the minor rant above is because I see in your post the same kind of misunderstanding that leads to people thinking that open source licenses are the same as being in the public domain. Even things like the BSD or MIT licenses, much less the GPL and its bretheren, maintain more control than public domain.
For the record, there *is* a music subscription service analogous to (if not quite as nice as) Netflix. For $15/month (less for long-term subscriptions), you can get a Zune Pass which includes all-you-can-eat music downloads. They're DRMed, of course, but the bitrate is respectable. The problem is that they won't play on anything but a Microsoft platform.
You also get 10 credits a month for free MP3 downloads, 320kbps. These are DRM-free, play in anything, and are usable on some music where MS hasn't been able to negotiate for Zune Pass streaming/downloads. If you consider a tyical cost of $1/song, those 10 credits make up 2/3 the cost of the monthly subscription by themselves.
Mind you, I think that's a stupid law. The maximum damage allowed should perhaps be relative to the crime. For example, somebody pirating a texting app should pay for a bunch of texts, as opposed to somebody pirating a texting app that lets them see their surroundings instead broadcasting a beacon to someobdy who runs them over with a car. However, I think in principle the idea of prohibiting this kind of thing is wrong. Impose whatever penalty you want on somebody who harms an innocent through such a trap, of course. That alone will keep the more excessive responses at bay. I just don't understand why somebody willfully violating the law should receive equal protection under it, though. Again, that's not to say they should receive *no* protection, but if I want to dig 10' pit traps to deal wiith those pesky kids trespassing on my lawn, I think I should have that right so long as the limits of my lawn are clearly delineated, and the traps aren't likely to cause extreme harm.
Windows Phone 7 does not allow side-loading by default, but there are several ways to enable it anyhow. The most legit option is to get a marketplace developer account, which costs a bit of money but also gives the right to publish on the marketplace. At the other end, there's ChevronWP7, which is still available on phone hacker sites. In between there are things like using LG's built-in registry editor to tweak the "allow sideload" option.
There is, for example, a NES emulator which is not in the WP7 marketplace but can be downloaded online and sideloaded.
Why spend all that time configuring when a clean install (of newer Windows versions) is about as fast, gets rid of *all* the crap as opposed to just the stuff you find, and gives you the disk space back as well?
Fine, so they move the keylogger into the BIOS/EFI. If you can't trust the hardware manufacturer (and clearly you can't, in the case of Samsung), there's nothing to do but examine every single component with an electron microscope to ensure that the RAM controller isn't shunting off everything coming from the keyboard buffer into a stream to the network card. I could write a rootkit that, installed on a hard disk's firmware microcontroller, is completely undetectable from the outside without physically removing the disk platters. If it sent info you might be able to intercept that info, but if it instead just did something like passively wait for a given date and then crash unrecoverably, you could (for example) bring down entire datacenters simultaneously - and there'd be no way a priori to know it was coming unless you directly examined the firmware (and not just what the FW reported itself to be, but the actual bits in the EEPROM or flash or whatever).
There are even several legit places to get the install DVD image. It's the license key that MS is really concerned about, not the bits. You can also use somebody else's DVD just fine, even if it's for a different edition (so long as it's for the correct architecture).
Right, because everybody knows that 90% of a game's code is in its UI and input system. Things like the game engine, AI, logic controlling elements in the game, resources, and netcode are completely irrelevant, right?
To be fair, WP7 doesn't support much in the way of netcode right now, and it's certainly not trivial to shift UI paradigms. However, that doesn't mean that the ability to use XNA, and resuse a lot of code as a result, isn't still quite valuable.
To provide some specific counter-points:
1) .NET is a framework, designed to provide a rich set of libraries that work with a number of front-end languages and produce programs that are portable across platforms, without even recompiling. The only sense in which it differs from the original goal of Java is that it emphasized cross-language more than cross-platform, although both frameworks support multiple front-end languages and run on multiple platforms today. The biggest implementation difference between the two, framework-wise, is that .NET allows lower-level access (direct "unsafe" pointers) than Java, and was designed from the start to avoid some of the pitfalls Java encountered around things like generics.
2) .NET allows *far* more versatility than most frameworks. It's not quite as good for metaprogramming as C++, or for data/code equivalence as Lisp or even Javascript, but it can get there easier than you might think. In the meantime, it supports everything from pointer access to functional programming, and you can mix a project written in C# that uses unsafe blocks to talk directly to driver code with one written in F# that has no side effects at all. You can use LINQ to easily and efficiently work with data collections in all kinds of interesting ways, Reflection to pull apart compiled assemblies and work with their parts directly, Attributes to easily extend the way that languages work... Hell, I'd say somebody who knows *all* the ins-and-outs of .NET probably knows more of how to solve any arbitrary problem than somebody who fully knows Java, C++, and LISP, and it's not only because the first guy actually knows *more* languages. I'd give the second guy the edge if he also knows at least two assembly languages.
Can you even read more than three sentences at once? His (Her?) issue isn't that Apple controls what ehy allow to be sold in their store, it's that Apple doesn't allow anybody to set up a competing store. If I wanted strippers I wouldn't go to a grocery store, I'd go to a strip club. If somebody tried to say "No, you can *only* buy things that the grocery store sells" then damn straight I'd protest (boycotting being a common choice of protest).
I take it you haven't tried BitLocker...
When you enable BL, the OS will generate a recovery key (a whole bunch of random alphanumerics; I think it actually has more entropy than the crypto key) and force you to store it somehow. Options include printing and writing to a flashdrive or other external storage (on a domain, it can be backed up by the domain controller). This recovery key works for any situation where the "easy" unlocking methods that you're using (for example, a TPM + smart card) don't work. It can be used to unlock or fully decrypt the volume on another computer, in case something goes drastically wrong.
What version of IE are you using that doesn't break those apps, but IE9 does? IE9 offers compatibility modes for 7 and 8 (the versions that shipped with the only OS versions which can run 9) and a "Quirks" mode that works with nearly all legacy sites designed for IE6 or older. We have no lack of stuff on the Intranet that either use corporate ActiveX controls or are desktop apps which embed MSHTML, and I think there's only one of them (which I've never used) that doesn't work correctly with IE9. I've been using trouble-free for a few weeks (RC) at work.
Customer Satisfaction results seemed to be very positive. http://www.bloomberg.com/news/2011-01-26/microsoft-says-it-shipped-2-million-windows-phones-last-quarter.html Details are still a little sketchy, but as of January it seems to be "well-regarded" by most users.
T-Mobile? What are you talking about? T-Mobile US carries the HTC HD7 and Dell Venue Pro, both WP7 devices, and has since the US launch day. They don't sell the DVP in their own stores, true, but they certainly do sell the HD7.
They don't advertise as heavily as AT&T, but it's certainly not true that they're only releasing this year.
Same in IE9. Tools -> Safety -> ActiveX Filtering. Blocks everything, including Flash and Java, just fine. It can be toggled on and off for individual sites quite easily; the little circle-with-a-line-through-it icon in the location bar (between the Search drop-down and the Compatibility Mode toggle button) works as a button to control it, and turns blue when filtering anything.