Not only did Red Hat's Szulik say "I can't answer that question" about whether he was in negotiations with Microsoft, but Microsoft "couldn't be reached for a comment" according to TFA. So, it is a safe bet there are discussions.
But that shouldn't be surprising. Red Hat said, after all, that
"We continue to believe that open source and the innovation it represents should not be subject to an unsubstantiated tax that lacks transparency."
- they never said "no deals with Microsoft about patents." And rightly so - the problem with the Novell-Microsoft deal is exactly that it is a vague 'innovation tax', mentions no actual patents, and is therefore completely nontransparent. Other deals regarding patents could be different.
A patent deal in which, say, Red Hat license specific WMA-related patents in order to ship WMA-playing software legally is 'reasonable' - at least assuming that that particular software is not GPL (or, if it is, if the patent license applies to all users of the software, not just Red Hat customers). I say 'reasonable' using scare quotes because this is reasonable only under the current highly-problematic US patent system. But, given that system, if Red Hat want to ship certain products, they must reach agreements concerning their patent holders. A sad fact of life.
(Note: I have no idea if WMA is patented, or not. This is just an example for illustration purposes.)
Microsoft's profits have dwarfed those of Sun and Apple combined and have relied on NOT [selling PCs themselves].
That is the generally assumed truth. Yet, I am starting to doubt it in recent years, and perhaps so does Microsoft.
The tipping point was the XBOX. Previously, Microsoft were adamant on their 'software-only' position (even in the portable music player market, which left it wide open for Apple, in retrospect). Then with the XBOX they changed that, and apparently it is starting to work well. The Zune is more along the same lines. While not successful, it shows Microsoft's intention to put out complete products, hardware and software combined.
But of course PCs are an established market, and Microsoft has a clear position in that market. But it is a dominant position, hardly vulnerable. If they sell PCs, what are their partners going to do, stop selling Windows on their computers? Not going to happen. So, in a sense, being software-only was what got Microsoft where they are (by letting hardware vendors fight amongst themselves and drop prices accordingly), but once established as basically the only operating system vendor for PCs, Microsoft can then branch off into selling PCs, and capturing even more profits.
There is really no reason why not, except potential antitrust litigation (but that seems less of a bother for them, recently).
"And, in fact, if the results were not statistically significant, they wouldn't get published very easily, and certainly not in Science."
Nonsense, much that is nonsense is published CERTAINLY in science, todays science is tomorrows superstition when dealing with crude measuring apparatus -- that being ultimately the human being which is prone to bias and overstating their interpretation of the data or doing folloup studies or later finding flaws in methodology that are not apparent, etc. Science is not immune from human frailty. One only has to look back 50-100 years to see how "scientific" many men were Freud, et, al.
Notice that I capitalized "Science". By that I meant the journal Science, not the human endeavor as a whole. Science, if you are not aware, is one of the 2 major empirical science journals (the other being Nature). Publishing there is among the highest achievements for scientists.
Indeed, scientific knowledge is prone to error. But, you cannot publish in the particular journal 'Science' if your results are not statistically significant - in the technical meaning of the term. Which is all I said, and all I meant. I did not say that the results were correct, just that they were statistically significant, which was doubted in the comment I was replying to. (You can certainly doubt if statistical significance leads to 'truth'; many scientists do.)
A difference of three IQ points seems almost within the margin of error
Since you say 'seems', I presume you didn't read TFA, otherwise you would know whether it did or didn't fall within the margin of error. And therefore it appears that you don't understand the concept of 'margin of error'. The margin of error can be arbitrarily small, it depends on the sample size.
In this study, they had 241,310 subjects. If memory serves me right, the population standard deviation is 15 points, so we have a margin or error along the order of 15 divided by the square root of 241,310, or 0.03. That is, two orders of magnitude smaller than 3 IQ points, which to you 'seems almost within the margin of error'.
Of course, the actual margin of error depends on other things, such as how many children were firstborn in the sample, how many were secondborn, etc. Still, with such a large sample, the final standard deviation should be much smaller than a single IQ point, making their conclusions statistically interesting. And, in fact, if the results were not statistically significant, they wouldn't get published very easily, and certainly not in Science.
They get acknowledgement for their patents. They have proof that people settled for their patents they can show in lawsuits down the line. Now their patents have assigned some value
I don't think so. The only thing acknowledged is that Linux vendors are dealing with Microsoft regarding patent issues. You might say that this shows, in some vague way, that Linux vendors believe Microsoft has patents that might be relevant. But which ones are they? A lawsuit would have to specify; but the Microsoft-Linux vendor deals don't mention specific patents, so they can't be used as 'precedent' (even in the nonlegal meaning of the word).
Sure, *most* people will surf porn, just as *most* people will watch American Idol, not PBS. Lowest common denominator and whatnot.
Minor correction. Most people watch American Idol, while some *other* people watch PBS. However, most people will surf porn, but this is *not* a distinct group than people visiting 'cultured' websites. Few people watch both American Idol and PBS; lots of people watch both porn and 'cultured' websites (albeit not at the same time, usually).
In other words, seeing lots of people visiting porn websites doesn't mean much about the overall level of the websites they visit. Whereas knowing that most (US) people watch American Idol says something quite sad.
What utter FUD. One does not sue the users of patent infringing software - one sues the makers of it.
No, one can sue both the makers and the users. Both are in infringement of the patent.
Sure, in a perfect world there wouldn't be software patents, or in a reasonable world you could only sue the makers about them, but in our actual world, users have gotten sued. Typically the manufacturer then gets involved (as happened with the Microsoft-Lucent case, IIRC).
Actually, had you bothered to read TFA, you would see that there is no flaming at all in Shwartz's post. He even invites Linus to dinner at the end. I am no fan of sun (although since GPLing Java I am starting to lean that way, I admit), but his reasoning in the post for several things (licensing choices of Solaris, relationship to Linux, etc.) makes a lot of sense.
Sure, we may see a nice flamewar here on Slashdot. But Sun, for their part, are not playing into that in any way. Actually even Linus's post was fairly tame (by Linus standards at least, he mentioned that he could be wrong about some things).
You can have closed source, statically linked application that uses Qt. Opera is a good example of this kind of application. You just need to get Qt with a correct licence and statically link to that.
Of course you are right, IIRC Skype is another example. But it costs money, that is, they expect something back if you develop on their platform (with Qt you can either be GPL, or pay Trolltech). With GTK+, on the other hand, you don't have these issues. Note that I'm not saying what is better (perhaps it is better to discourage proprietary development?), just facts.
I do find the notion of taking an open source project, slapping on a bit of code, and trying to keep your part to yourself, to be extremely offensive. The people who contributed to GNU software did so with the expectation that their code is share and share alike.
In some way, yes, they did expect 'share and share alike'. But not entirely. Linux kernel developers explicitly allow you to do whatever you want in userspace; they don't expect anything from you if you use their kernel and 'slap' a proprietary app on top of it. Likewise GTK+ devs don't expect anything if all you do is use GTK+ as your GUI toolkit, i.e., just link to it externally. And in fact a lot of proprietary development prefers GTK+ (which is LGPL in the relevant part) over Qt, which is more problematic for them.
In fact this is the explicit reason for the LGPL being different from the GPL, and similar things (GPL+exceptions, etc.). To return to the GNOME/GTK+ example, the (GNU!) licensing allows proprietary apps to run on their platform; in that sense this is a very business-friendly license (to FOSS and proprietary businesses alike). And this makes sense: even if apps developed on top of GNOME/GTK+ are proprietary, often during their development some contribution is made back to GNOME/GTK+, in the form of bug reports, patches, and perhaps other things; in addition, if the app is widely-used then it gets more people familiar with the platform underneath it, opening more potential avenues for benefit to GNOME/GTK+ (and of course this example is relevant to other projects as well: the Linux kernel, languages/runtimes like Python, etc.).
As long as you link externally to LGPL code, or only write in userspace while using something like the Linux kernel, you have no problem. Between these two they allow basically everything you need (assuming that in fact you don't need to write your own kernel modules or make changes inside the LGPL code).
This is precisely how proprietary apps work on Linux (examples: Matlab, Oracle, etc. etc.). In fact you can even use a GUI, assuming you pick GNOME/GTK+ (not KDE/Qt).
So, basically you have no problems at all. Yes, BSD is somewhat simpler in that you don't need to check these issues. But really there is no reason not to use Linux (if it suits your goals, of course, BSD is also very good).
Since you intend to give out boxes including the Linux kernel, etc., you will need to abide by the GPL, i.e., provide source for for the kernel, etc., NOT for your own code. Note, however, that from GPL3 and onwards, assuming projects adopt it, you will not be able to Tivo-ise your hardware. If you run in userspace anyhow, this probably won't be an issue, but you do need to consider this beforehand.
What is certainly good advice is the following. Don't get too tied to a single platform! Write your app as portable as possible, so that you can deploy it on Linux, BSD, OpenSolaris, etc. (and if you have a GUI, then using a cross-platform toolkit, and separating GUI from backend as much as possible to facilitate even changing that). Then should issues arise (not just with licensing!) you can adapt.
Question: Can Samsung/LG/etc legally use Linux in their product once
1) They have signed one of these satanic "patent deals"
2) GPL3 is out
We haven't seen the wording of the deal, so we can't say. Perhaps the agreement is so vague that it covers nothing that can be pinned down - like the Novell deal? Such deals appear to be (to me) just words, with no legal effect (if they wanted to protect actual products, they would name them... not too hard, is it?).
My personal take on all these deals is that Microsoft is aiming for one of the following:
1. Cause strife in Linuxland, perhaps via GPL3 matters, perhaps just by causing many in the community to hate Novell, etc. etc. Divide and conquer and all that. There seems to be some moderate success at this already. Not sure if it's worth hundreds of millions of dollars, though, so I doubt this is the whole picture (perhaps just a side benefit).
2. Cover Microsoft's ass in case Linux takes off. Sure, currently they pay more to LG, Novell, etc., because they ship more Windows than the other party ships their products. But, if Linux becomes the standard on (say) smartphones and other devices, then suddenly the tables may turn, and Microsoft may see money flowing the other way. In other words, if Microsoft stays on top, they pay a little pocket change; if Linux explodes, then at least Microsoft can comfort themselves with some patent royalties. That is a reasonable business strategy (carried out in an unethical manner, however, IMO).
It is thus logical that a truly superior human will learn to abandon any primitive altruistic tendencies.
If altruism is a natural inborn human quality, just like language, then perhaps a 'superior being' should abandon language as well? Why does abandoning one inborn quality seem 'superior', but another not? Why should altruism be 'primitive' but language not?
Is it because altruism appears to benefit others, and a 'superior being' would presumably do things that are better for himself? Well, that presumption is at fault; who says what is 'better'? If doing only for yourself is 'better', then perhaps you yourself have already abandoned morality.
In the end, what is 'superior' is a matter of opinion; there are no absolutes in this matter. That morality may be, in part, inborn does nothing to further that debate.
You make some very good points. Indeed, Genesis 1-3 really doesn't make too much sense when read directly. When contrasted with the myths of other peoples in the area at the time it does become a little clearer (e.g. the plural form, corresponding to polytheism; the assumptions of a specific morality, corresponding to the norms of the period, etc.). It's a fascinating cultural product, IMO.
What sin? She ate a fruit when it was offered to her, by a being that God _allowed_ into the Garden. Yes, she was told not to eat the fruit, but was never told why or what the consequences were, despite God being omniscient enough to know he had created man with curiosity.
I have to correct you here: God did tell them what the consequences were: he said that if they ate from the tree they would die. Pretty straightforward. Actually ironic though, because the end of the story of the Garden of Eden is with God (in the odd plural) saying that should Adam&Eve eat from the tree of Life (and not just of Knowledge) then they would live forever. So it seems that they would die anyhow, whether they eat from the tree of Knowledge or not.
Genesis 1 to 3 is really a peculiar thing to read.
That's true, but note that "Clone Products" in this contract are only with respect to the other party. So Linux implementing *NIX isn't an issue since *NIX isn't a Microsoft product.
I am no lawyer (but I do read contracts from time to time, as a 'hobby'), but this is really an odd 'covenant'. The agreement appears to not state what products are actually covered by the patent covenant, in bizarre ways. For example, "Clone Products" are not covered, "Clone Products" being presumably things like Mono and OpenOffice (as they duplicate Microsoft APIs and products); yet all such products already designed at time of signing are exempt, i.e., they are covered. Yet, the following projects are not subject to the exemption: "Wine, OpenXchange, StarOffice and OpenOffice", i.e., they are not covered. So OpenOffice appears to not be covered.
Likewise Samba would presumably be a "Clone Product", and not covered as well, except by the exemption due to its existing at time of signing. Yet this might not cover additional functionality added later. It just isn't clear.
No actual products are named aside from the quote above, and even they are not stated as being covered or not (just not exempted by a particular subsection). So, reading this, I can't tell whether Novell customers are in fact covered or not, in any way. The assumption was always that the agreement did protect them from patent lawsuits. But that assumption may have been wrong.
Is the contract specifically designed to not mention any products, effectively letting it be ambiguous and perhaps of no legal use - that is, only effective for PR purposes?
Comparing identical models is interesting, and it's good to know the size of the 'Windows Tax' is around $50 (as many suspected), but this figure isn't an objective truth, for several reasons:
Ubuntu can run well on cheaper hardware than Vista (mainly RAM and video, if you want Aero). So comparing the same hardware means one OS will run better than the other. (Note: on the other hand Ubuntu needs more expensive hardware in some areas, like Wifi, due to lack of good drivers for cheap Broadcom devices. But this is negligible, and also drivers are now coming out.)
Comparing to Vista Home Basic may not be entirely fair, if most users in fact purchase Home Premium ($30 more via Dell, I believe). Indeed some have said that Home Basic is hopelessly crippled. But this is of course debatable.
At least the advantage of Gaming is that there is no real inertia, everyone starts at the same point every five years. If MS gains a monopoly its usual tactics won't keep it there.
It does seem that way, doesn't it? Just remember, there was no OS monopoly before Windows, and no office monopoly before MS Office. Microsoft are experts at generating self-propagating monopolies (yes, they failed at many things - MSN, Zune - but a few big successes are all they need). Just watch, they are already leveraging their Windows monopoly to help the XBOX (multiplayer games with Windows and XBOX, shared credit system, etc.), they are consolidating their development tools so that developing for Windows and XBOX can be done simultaneously, and I am sure they have more tricks up their sleeve (look at the XBOX media store, I am sure we will see tight integration with Windows).
So yes, people buy a new console every 5 years. But only the Microsoft console will integrate especially well with their Windows machine, both for the end-user and the developer. We may very well see a console monopoly in the next decade. Sadly, Sony and Nintendo seem to have no clue (if they did, we would see them standardizing their development tools, integrating with iPods, etc.).
You can also BUY services through CNR. CNR is not only a package manager, it is also a store. If you pay for MP3 codecs through CNR, then that is legal.
Really? That is odd then. Over here, you can switch the language of Hebrew Windows to English, last I checked. But that was years ago, perhaps they changed it...
Yes, but if Dell intend people to break the (ridiculous, but existent) law to play MP3s, then that seems odd: doesn't that open Dell up to a lot of liability? (contributory infringement, that sort of thing)
The point isn't that CNR is easier, it is that CNR is legal. You can buy the proprietary codecs, etc., from there. Downloading them through "add/remove programs" is (sadly, stupidly) not legal in the US.
Slashdot Mirror
But that shouldn't be surprising. Red Hat said, after all, that - they never said "no deals with Microsoft about patents." And rightly so - the problem with the Novell-Microsoft deal is exactly that it is a vague 'innovation tax', mentions no actual patents, and is therefore completely nontransparent. Other deals regarding patents could be different.
A patent deal in which, say, Red Hat license specific WMA-related patents in order to ship WMA-playing software legally is 'reasonable' - at least assuming that that particular software is not GPL (or, if it is, if the patent license applies to all users of the software, not just Red Hat customers). I say 'reasonable' using scare quotes because this is reasonable only under the current highly-problematic US patent system. But, given that system, if Red Hat want to ship certain products, they must reach agreements concerning their patent holders. A sad fact of life.
(Note: I have no idea if WMA is patented, or not. This is just an example for illustration purposes.)
The tipping point was the XBOX. Previously, Microsoft were adamant on their 'software-only' position (even in the portable music player market, which left it wide open for Apple, in retrospect). Then with the XBOX they changed that, and apparently it is starting to work well. The Zune is more along the same lines. While not successful, it shows Microsoft's intention to put out complete products, hardware and software combined.
But of course PCs are an established market, and Microsoft has a clear position in that market. But it is a dominant position, hardly vulnerable. If they sell PCs, what are their partners going to do, stop selling Windows on their computers? Not going to happen. So, in a sense, being software-only was what got Microsoft where they are (by letting hardware vendors fight amongst themselves and drop prices accordingly), but once established as basically the only operating system vendor for PCs, Microsoft can then branch off into selling PCs, and capturing even more profits.
There is really no reason why not, except potential antitrust litigation (but that seems less of a bother for them, recently).
Of course, you are perfectly reasonable in considering 3 points to not be much, objectively speaking.
That is a completely separate issue from statistical significance, which is the topic I was referring to.
Indeed, scientific knowledge is prone to error. But, you cannot publish in the particular journal 'Science' if your results are not statistically significant - in the technical meaning of the term. Which is all I said, and all I meant. I did not say that the results were correct, just that they were statistically significant, which was doubted in the comment I was replying to. (You can certainly doubt if statistical significance leads to 'truth'; many scientists do.)
In this study, they had 241,310 subjects. If memory serves me right, the population standard deviation is 15 points, so we have a margin or error along the order of 15 divided by the square root of 241,310, or 0.03. That is, two orders of magnitude smaller than 3 IQ points, which to you 'seems almost within the margin of error'.
Of course, the actual margin of error depends on other things, such as how many children were firstborn in the sample, how many were secondborn, etc. Still, with such a large sample, the final standard deviation should be much smaller than a single IQ point, making their conclusions statistically interesting. And, in fact, if the results were not statistically significant, they wouldn't get published very easily, and certainly not in Science.
Minor correction. Most people watch American Idol, while some *other* people watch PBS. However, most people will surf porn, but this is *not* a distinct group than people visiting 'cultured' websites. Few people watch both American Idol and PBS; lots of people watch both porn and 'cultured' websites (albeit not at the same time, usually).
In other words, seeing lots of people visiting porn websites doesn't mean much about the overall level of the websites they visit. Whereas knowing that most (US) people watch American Idol says something quite sad.
Sure, in a perfect world there wouldn't be software patents, or in a reasonable world you could only sue the makers about them, but in our actual world, users have gotten sued. Typically the manufacturer then gets involved (as happened with the Microsoft-Lucent case, IIRC).
Actually, had you bothered to read TFA, you would see that there is no flaming at all in Shwartz's post. He even invites Linus to dinner at the end. I am no fan of sun (although since GPLing Java I am starting to lean that way, I admit), but his reasoning in the post for several things (licensing choices of Solaris, relationship to Linux, etc.) makes a lot of sense.
Sure, we may see a nice flamewar here on Slashdot. But Sun, for their part, are not playing into that in any way. Actually even Linus's post was fairly tame (by Linus standards at least, he mentioned that he could be wrong about some things).
In fact this is the explicit reason for the LGPL being different from the GPL, and similar things (GPL+exceptions, etc.). To return to the GNOME/GTK+ example, the (GNU!) licensing allows proprietary apps to run on their platform; in that sense this is a very business-friendly license (to FOSS and proprietary businesses alike). And this makes sense: even if apps developed on top of GNOME/GTK+ are proprietary, often during their development some contribution is made back to GNOME/GTK+, in the form of bug reports, patches, and perhaps other things; in addition, if the app is widely-used then it gets more people familiar with the platform underneath it, opening more potential avenues for benefit to GNOME/GTK+ (and of course this example is relevant to other projects as well: the Linux kernel, languages/runtimes like Python, etc.).
As long as you link externally to LGPL code, or only write in userspace while using something like the Linux kernel, you have no problem. Between these two they allow basically everything you need (assuming that in fact you don't need to write your own kernel modules or make changes inside the LGPL code).
This is precisely how proprietary apps work on Linux (examples: Matlab, Oracle, etc. etc.). In fact you can even use a GUI, assuming you pick GNOME/GTK+ (not KDE/Qt).
So, basically you have no problems at all. Yes, BSD is somewhat simpler in that you don't need to check these issues. But really there is no reason not to use Linux (if it suits your goals, of course, BSD is also very good).
Since you intend to give out boxes including the Linux kernel, etc., you will need to abide by the GPL, i.e., provide source for for the kernel, etc., NOT for your own code. Note, however, that from GPL3 and onwards, assuming projects adopt it, you will not be able to Tivo-ise your hardware. If you run in userspace anyhow, this probably won't be an issue, but you do need to consider this beforehand.
What is certainly good advice is the following. Don't get too tied to a single platform! Write your app as portable as possible, so that you can deploy it on Linux, BSD, OpenSolaris, etc. (and if you have a GUI, then using a cross-platform toolkit, and separating GUI from backend as much as possible to facilitate even changing that). Then should issues arise (not just with licensing!) you can adapt.
We haven't seen the wording of the deal, so we can't say. Perhaps the agreement is so vague that it covers nothing that can be pinned down - like the Novell deal? Such deals appear to be (to me) just words, with no legal effect (if they wanted to protect actual products, they would name them... not too hard, is it?).
My personal take on all these deals is that Microsoft is aiming for one of the following:
1. Cause strife in Linuxland, perhaps via GPL3 matters, perhaps just by causing many in the community to hate Novell, etc. etc. Divide and conquer and all that. There seems to be some moderate success at this already. Not sure if it's worth hundreds of millions of dollars, though, so I doubt this is the whole picture (perhaps just a side benefit).
2. Cover Microsoft's ass in case Linux takes off. Sure, currently they pay more to LG, Novell, etc., because they ship more Windows than the other party ships their products. But, if Linux becomes the standard on (say) smartphones and other devices, then suddenly the tables may turn, and Microsoft may see money flowing the other way. In other words, if Microsoft stays on top, they pay a little pocket change; if Linux explodes, then at least Microsoft can comfort themselves with some patent royalties. That is a reasonable business strategy (carried out in an unethical manner, however, IMO).
Is it because altruism appears to benefit others, and a 'superior being' would presumably do things that are better for himself? Well, that presumption is at fault; who says what is 'better'? If doing only for yourself is 'better', then perhaps you yourself have already abandoned morality.
In the end, what is 'superior' is a matter of opinion; there are no absolutes in this matter. That morality may be, in part, inborn does nothing to further that debate.
You make some very good points. Indeed, Genesis 1-3 really doesn't make too much sense when read directly. When contrasted with the myths of other peoples in the area at the time it does become a little clearer (e.g. the plural form, corresponding to polytheism; the assumptions of a specific morality, corresponding to the norms of the period, etc.). It's a fascinating cultural product, IMO.
Genesis 1 to 3 is really a peculiar thing to read.
That's true, but note that "Clone Products" in this contract are only with respect to the other party. So Linux implementing *NIX isn't an issue since *NIX isn't a Microsoft product.
Link to actual agreement
I am no lawyer (but I do read contracts from time to time, as a 'hobby'), but this is really an odd 'covenant'. The agreement appears to not state what products are actually covered by the patent covenant, in bizarre ways. For example, "Clone Products" are not covered, "Clone Products" being presumably things like Mono and OpenOffice (as they duplicate Microsoft APIs and products); yet all such products already designed at time of signing are exempt, i.e., they are covered. Yet, the following projects are not subject to the exemption: "Wine, OpenXchange, StarOffice and OpenOffice", i.e., they are not covered. So OpenOffice appears to not be covered.
Likewise Samba would presumably be a "Clone Product", and not covered as well, except by the exemption due to its existing at time of signing. Yet this might not cover additional functionality added later. It just isn't clear.
No actual products are named aside from the quote above, and even they are not stated as being covered or not (just not exempted by a particular subsection). So, reading this, I can't tell whether Novell customers are in fact covered or not, in any way. The assumption was always that the agreement did protect them from patent lawsuits. But that assumption may have been wrong.
Is the contract specifically designed to not mention any products, effectively letting it be ambiguous and perhaps of no legal use - that is, only effective for PR purposes?
So yes, people buy a new console every 5 years. But only the Microsoft console will integrate especially well with their Windows machine, both for the end-user and the developer. We may very well see a console monopoly in the next decade. Sadly, Sony and Nintendo seem to have no clue (if they did, we would see them standardizing their development tools, integrating with iPods, etc.).
You can also BUY services through CNR. CNR is not only a package manager, it is also a store. If you pay for MP3 codecs through CNR, then that is legal.
Really? That is odd then. Over here, you can switch the language of Hebrew Windows to English, last I checked. But that was years ago, perhaps they changed it...
Yes, but if Dell intend people to break the (ridiculous, but existent) law to play MP3s, then that seems odd: doesn't that open Dell up to a lot of liability? (contributory infringement, that sort of thing)
The point isn't that CNR is easier, it is that CNR is legal. You can buy the proprietary codecs, etc., from there. Downloading them through "add/remove programs" is (sadly, stupidly) not legal in the US.