No, it's a syntax error. "The other 95% of programmers" refers to the complete set of programmers, less excluded subset. He defined that subset as himself, instead of himself plus others who can code properly; improper usage of "The other" is what caused his dependent clause to be false.
Its not a syntax error: the sentence is grammatically correct but does not have the intended semantics. Its a "bug": if it were code, it would compile and run, but just produce results other than what was intended.
Eliminating a class of bugs is a great thing, but GC doesn't do it.
Yes, it does. As you admit in the next sentence.
Instead of memory leaks, you have object leaks.
Which is a completely different problem, even if it can be produced by overlooking the same type of things. And while it may create performance problems, it is less likely to create remotely exploitable vulnerabilities.
People start to trust the GC too much, and it takes away more control than it's worth.
SUre, anything that makes things easier in most cases can be a source of overreliance that makes the edge cases where it doesn't relieve the need for attention. But most modern GCs don't take away that much control. You may lose control by choosing not to worry and to just trust the GC, but I suspect that most people that do that do so because they work in environments where, overall, it is worth it.
I've said it before and I'll say it again: the OLPC will end with a massive taxpayer bailout when congress is asked to spend $300 million on laptops that no one wants.
Why would Congress bail out either the asian manufacturer or the nonprofit? Congress' boondoggle bailouts are generally restricted to for-profit US firms.
Not to mention, the laptop's yet to be piloted in any real-world implementations, so wherever this money's coming from, it may not really be going anywhere.
While it hasn't been used on a large scale, it has been piloted in real-world, classroom implementations in several of the participant countries (including, at a minimum, Brazil, Peru, Uruguay, Nigeria, and Thailand [the last of which backed out of the project after the recent coup]).
So how well does this work when someone is living in a grass hut in Africa and they do now have electricity.
I assume you mean "not" instead of "now". In any case, people living in rural villages without regular access to electricity is exactly why the OLPC is designed to use very little power, and to be manually charged.
I guess they could eat the laptop because we all know a laptop is better then food.
Well, yes, as an educational tool for people who aren't starving, a laptop is better than food.
Also what is the benefit of this? A child has a lap top to use and learn on. My computer has never taught me anything just because it is on and running Linux or Window.
The software and educational content supplied with the XO and developed by the OLPC project is not limited to the operating system (though certainly the desktop environment itself is designed to facilitate learning, though not to "teach" on its own.) But, mostly, its a tool for the education ministries to use to enable them to deliver the content they want and conduct the educational activities they prefer.
It is the power to connect to the internet, download compilers and execute programs and do research online that has taught me. Will these children have the internet in there area or is this just a ti graphing calculator with a bigger keyboard and a few applications.
One of the things developed by partners in the OLPC project and being made available for purchase to the participants is a satellite downlink station designed for rural villages; the same provider is donating satellite time to provide internet connections to remote locations. The OLPC project itself will also be making available school-site servers to provide access points, and the laptops themselves can form ad hoc networks. So, yes, the internet will be available (though perhaps not "always on" for most users), and ad hoc networking and other forms of interaction will also be available.
Why not take the $100 and go to that child's town and make a change.
Presumably, the national governments that are the only people spending the cost of the laptop (which is >$100), are generally already spending far more than that on other projects. If they expected to get more marginal value by spending money in other improvements than on the laptops, they wouldn't be buying the laptops. Of course, if you have specific complaints about how particular of the governments should spend their money better, feel free to make them.
Say you have 100 children in a town so, $10,000, take that money and help them redesign the town so crops can grow better.
Presumably, if the national governments thought they had a way to spend that kind of money that would substantially improve agricultural output (especially in the many OLPC countries where that's the main industry), they would be doing that, whether or not they were buying the laptops. But, if you see areas where particular countries that are purchasing the OLPC are missing opportunities in this regard, feel free to point them out.
If someone handed me the controls to a Nuclear Reactor I would not know how to use it, much like if you hand a child a computer and the only toy they have is a bike from the 1970's.
Presumably, governments that are going to be spending, to meet the one-per-school-age-child target, billions of dollars on just the OLPC hardware and that have already had students, teachers, and education ministry staff involved in demonstration projects are going to roll the laptop out with something a little more sophisticated that dumping them on students with no instruction; the whole idea has been to integrate them into the curriculum. Of course, if you know of specific countries that have announced the intention to purchase the laptops and just dump them on school age children, please, feel free to point to them.
Most of these people would probably better off getting $100 which is like 5 years wages where they live.
In what is, IIRC, the largest launch country, Brazil, median income for black women (the worst off racial/gender mix) is $156/month. (source)
Heck, even Rwanda (which is one of the poorest nations that may get it early, through Libya purchasing it for them) has an average per capita annual income of $206 (source), over an order of magnitude higher than you suggested for "most" OLPC recipients.
Will kids in the states also be eligible for these?
There is no such thing as individual "eligibility" for the laptops, so the question is incoherent. Yes, the US Department of Education is as free as any other national education ministry to purchase the laptops for distribution on a one-per-child basis, though of course they aren't the principal target market and the OLPC feature set is designed around use in a very different environment than one of the most developed nations in the world.
I am also saying that no high level language is a good substitution for good programming practices.
In many situations, higher-level languages reduce the cost of "good programming practices" by reducing the number of specific concerns that need to be addressed in a program designed for a given task, compared to C or C++.
They are not intended as a substitute for good practice, nor has anyone that I have seen suggest them as a substitute.
Garbage collected languages is no solution to poor programming.
Eliminating the possibility of an entire class of bugs reduces the number of things that need to be remembered in putting the code together (and checked once it is coded), and therefore reduce the number of bugs (and exploitable vulnerabilities) that get through to end user code.
Having an ODF standard should not exclude an OOXML standard.
There are two arguments here: The first is that, independently of the existence of any other standards covering the same subject matter, OOXML is a poorly described, non-implementable, and otherwise bad proposed standard, and should be rejected. The second is that that the existence of one standard covering a topic makes additional standards covering the topic less valuable, potentially redundant, and in some cases contradictory to the purpose of standardization, particularly when adopted by the same standards body.
Debate over the second seems to only make sense in a context where it is assumed or concluded that OOXML would be a desirable standard on its own in the first place.
As was pointed out to you already. TiVo is free to use the GPL'd software for the purpose they are using it now - PVR. What they won't be able to do anymore is to remove "our" right to modify that same software.
Tivo doesn't remove your right to modify the software now. The source is available. You can modify it. What they don't do is give you a system that will run modified software, nor an easy way to produce one. There are a few ways the FSF could have addressed that, presuming that concern is important (to me, it seems somewhat out-of-scope of the focus of the pre-v3 GPL, but it is a concern). One of them would have been to require equivalent openness of hardware: hardware that bundles GPL software would be required to have open specifications with rights that enable the recipient of the bundled hardware/software to "rebuild" their own implementation of the hardware (either a faithful one or a modified copy) and run whatever software they want on it. This would have allowed users freedom to do what they want, as well freedom to select hardware/software-bundle features.
Another approach, and the one the FSF chose—but only for consumers, because of outcry that business users needed the security of locked-down hardware/software bundles—was to have prohibit locked-down hardware. This requires it to be cheaper for users to run modified software than it might be under the other model, since they won't have to "rebuild" hardware, at the expense of the freedom to choose features of a hardware/software bundle they might prefer.
Woah - TiVo isn't using that software, their customers are using the software. The FSF is telling Tivo, if you're giving our software to your users, you have to give them the ability to change it.
Or, looked at a different but equally accurate way, it is telling them (but not people who sell hardware with GPLv3 software that exclusively targets "business" users), "if you're giving our software to your users, you must not give them the option of hardware that provides an absolute guarantee against changes". IOW, it limits customer freedom to choose hardware features that they might be interested in.
Sure, hackers and tinkerers who are pretty certain that they'll be the most technically proficient users with physical access to their home hardware probably don't want that feature. Technophobic parents whose children are (or who perceive their children to be) far more technically proficient, OTOH, might prefer that feature. However, the FSF has declared that while businesses might have a need for hardware with that feature that makes it important to keep that choice available, consumers can't have that feature if they want OSS. In fact, it so important to keep that feature away from consumers that products that are "dual use" (marketed both to businesses and consumers) can't have the feature, only "business-only" products.
And if your business-only product is designed to be incorporated into a dwelling (say, you make a digital utility meter), well, then its a consumer product, and you can't have that feature, either.
Oh dear lord. Email is not ruled by the same laws governing the USPS. There is no mail fraud here people!
If Microsoft, like many other online service providers, advertises or solicits business via the mail (certainly, they've done that for MSN, though I don't know if they have for Hotmail per se), it is governed by the same law that governs anyone else making such solicitations (not the USPS, but other postal service users).
OTOH, any online fraudulent solicitations by Microsoft would be more likely to be wire fraud, but Microsoft may be insulated from such charges from "free" users since Microsoft, while it uses them to get money from advertisers who hope to target them, does not get money or property from the users directly.
On the third hand, depending on how they market to advertisers, they may be guilty of fraud (regular, wire, mail, or all three) if they've misrepresented to them the kind of service their advertising will be associated with, since that is quite arguably a material misrepresentation directly to induce the advertiser to give money or property to Microsoft.
Your anti-war demonstration scenario is only going to get your property frozen if it's a violent demonstration.
Or, rather, if executive branch officials without outside review decide that your demonstration suggests some potential future inclination toward violence.
And, actually, it's not 5th amendment, which is the right to due process, but the 4th amendment, which is the right against unwarrantable searches and seizures, that is violated.
How is it not a deprivation of property without due process?
Why is it so hard to get a computer to do something that a die can do easily?
Because the (approximate, though unfailingly somewhat biased) randomness of a die relies on uncertainties in physical conditions in the roll, whereas computers are designed to work predictably, to produce the same output to a given input. Even if you built a physical die roller into a computer system, I suspect it would be a lot of work to make it is good a random number generator as a human repeatedly rolling the same die.
Just because an argument's ancient doesn't mean it's not still valid. Plus, after all, the number of distributions has been rising for a long time. Maybe the argument carries more weight now than it used to.
Seems to me, Linux adoption has grown over the same time. So, if anything, the argument carries less weight now.
Well, if the Linux community had the first clue about how to market Linux to the average mouthbreather
Unfortunately, the Linux (and open source more generally) community is, as a broad generality, much better at creating functioning software (particularly the parts "behind" the UI) than marketing or, to a lesser extent, making well-polished chrome; whereas the Windows "Community" (i.e., Microsoft) is much better at marketing and making chrome than building solid software internals.
It seems to me that "Loose" sounds like "Loose"—if you take most words in English that end with "-oose" as a guide, e.g., "moose", "goose", or "caboose".
OTOH, "lose" sounds like "Loose"—if you take "choose" as a model of how the "-oose" should sound.
I think the better statement is that "English spelling is only loosely connected to pronunciation".
I am wondering if this information may or may not discount the theory the Homosapians and Neanderthalls in Europe may have cross breaded?
Whether H. sapiens and H. neanderthalensis interbred to result in H. sapiens sapiens or whether, OTOH, H. sapiens sapiens evolved from H. sapiens alone before, after, or while the latter displaced H. neanderthalensis in Europe is somewhat irrelevant to the question of whether H. sapiens had a single origin in Africa.
It might be relevant to the question of whether H. sapiens sapiens had a single origin, which would be a different question. (A subspecies without a single origin would seem to be less surprising; a species with multiple origins would be, AFAIK, rather unusual.)
and yet many OSS advocates plug OSS as the be all and end all.
So?
you can't have it both ways.
Sure you can, because there aren't two different "ways" at issue for the same thing.
Sure, die-hard OSS advocates think that all software should be available under an OSS licensing model, both free and gratis.
That does not mean that they think that enterprises should just use the software and expect someone else to develope it to their needs and rely on free community support. Many OSS advocates are completely behind the idea of enterprises paying for support and development of OSS software (in fact, that's how many OSS advocates manage to eat) to meet their needs rather than paying for support and licensing of commercial software, selling as an advantage that, should their support vendor collapse for one reason or another, it will be easier for the enterprise to spend the same money to get the support for the software from another vendor or to develop support capacity in house than with commercial software where IP licensing concerns may make that more difficult to arrange without changing software.
Selling OSS as the ideal licensing approach doesn't mean that OSS is a free lunch for business users in terms of support. What it does mean is that the vendor choosing to go a different direction or being purchased by a competitor won't ever make support legally impossible. But nothing takes out the requirement that, for support to happen, there has to be someone willing to do the support, and that money will often be an important motivating factor in making that happen.
When people make statements that are obviously wrong but stated politely, must we still be polite in return?
I don't know, do you think people should be polite to you or not?
First, you are wrong when you state that one could take Tivo's code, modify it, and use the modified software on one's own hardware.
No, this is completely true, and nothing you say "supporting" your argument that this is "wrong" even contradicts it, much less proves it wrong.
When one purchases a Tivo DVR, that device now belongs to the purchaser, not to the seller, and not to the manufacturer.
This is true, of course, but doesn't contradict the idea that you can take the code, modify it, and use the modified software on one's own hardware.
The DRM hardware prevents the new owner from running his choice of operating systems and applications on his own Tivo DVR device.
It also prevents people who are not the owner who get access to the device from running their choice of software on the Tivo DVR device. As the authors of the GPLv3 apparently recognized, the freedom to choose systems with this feature is important to some users. However, unfortunately, the GPLv3 authors decided that the freedom to include such a feature must be denied to consumers but preserved for business, with the preference for the former so strong that if a product has both consumer and business use, denying the freedom for consumers to select a product with such a feature overrides preserving the option for business.
Second, you are wrong when you state that software licensed under the GPL3 could not be used in voting machines.
That's possibly true, though the combination of the vague definition of "user product" in section 6 and the various terms that push all questionable or mixed cases into that category may create enough of a legal risk to prevent the use of such software in voting machines.
The GPLv3, whether one approves of it or not, is a sharp departure in that it incorporates restrictions on product features, which the GPLv2 did not, and that it has terms (those same restrictions) which substantially vary in effect based on who a product is marketed to and what is marketed for, whereas the GPLv2 had the same substantial effect no matter who a product was targetted at or how it was to be used.
Unlike the GPLv2, the GPLv3 is based, it appears from its substantive terms, not on the idea of a single, common set of "software freedoms", but on the idea that the FSF knows best which users ought to have which freedoms.
You obviously don't go to church. If you go to a church where the preacher consistently puts forward a version of the faith that you disagree with, you go to another church.
The FSF may or may not be analogous to a "church" with a "faith". But, AFAIK, Linus isn't a member of the FSF, and has never claimed to adhere to its "faith".
The GPLv2 isn't a church. Its a license. It has concrete terms. People (at least, some of them) decide to use it, or not, based on how well they determine those terms are aligned with their desires for their software. Linus, I expect, uses the GPLv2 for that reason, not out of quasi-religious faith in the cult of RMS.
In any case, so long as we're in agreement that Linus doesn't believe in freedom or defending it
I think Linus believes in freedom. I don't think he agrees with the FSF's understanding of what freedom is and how it should be defended as expressed in the GPLv3, though. Heck, I think that Linus's differences with the FSF on those points predate the GPLv3 discussion by quite a bit, though the GPLv2 was something he could live with though he understood some of its terms differently from the FSF.
Slashdot Mirror
Coming in 2008:
Microsoft Office Live! Ballmer Vista
Its not a syntax error: the sentence is grammatically correct but does not have the intended semantics. Its a "bug": if it were code, it would compile and run, but just produce results other than what was intended.
Yes, it does. As you admit in the next sentence.
Which is a completely different problem, even if it can be produced by overlooking the same type of things. And while it may create performance problems, it is less likely to create remotely exploitable vulnerabilities.
SUre, anything that makes things easier in most cases can be a source of overreliance that makes the edge cases where it doesn't relieve the need for attention. But most modern GCs don't take away that much control. You may lose control by choosing not to worry and to just trust the GC, but I suspect that most people that do that do so because they work in environments where, overall, it is worth it.
Why would Congress bail out either the asian manufacturer or the nonprofit? Congress' boondoggle bailouts are generally restricted to for-profit US firms.
While it hasn't been used on a large scale, it has been piloted in real-world, classroom implementations in several of the participant countries (including, at a minimum, Brazil, Peru, Uruguay, Nigeria, and Thailand [the last of which backed out of the project after the recent coup]).
I assume you mean "not" instead of "now". In any case, people living in rural villages without regular access to electricity is exactly why the OLPC is designed to use very little power, and to be manually charged.
Well, yes, as an educational tool for people who aren't starving, a laptop is better than food.
The software and educational content supplied with the XO and developed by the OLPC project is not limited to the operating system (though certainly the desktop environment itself is designed to facilitate learning, though not to "teach" on its own.) But, mostly, its a tool for the education ministries to use to enable them to deliver the content they want and conduct the educational activities they prefer.
One of the things developed by partners in the OLPC project and being made available for purchase to the participants is a satellite downlink station designed for rural villages; the same provider is donating satellite time to provide internet connections to remote locations. The OLPC project itself will also be making available school-site servers to provide access points, and the laptops themselves can form ad hoc networks. So, yes, the internet will be available (though perhaps not "always on" for most users), and ad hoc networking and other forms of interaction will also be available.
Presumably, the national governments that are the only people spending the cost of the laptop (which is >$100), are generally already spending far more than that on other projects. If they expected to get more marginal value by spending money in other improvements than on the laptops, they wouldn't be buying the laptops. Of course, if you have specific complaints about how particular of the governments should spend their money better, feel free to make them.
Presumably, if the national governments thought they had a way to spend that kind of money that would substantially improve agricultural output (especially in the many OLPC countries where that's the main industry), they would be doing that, whether or not they were buying the laptops. But, if you see areas where particular countries that are purchasing the OLPC are missing opportunities in this regard, feel free to point them out.
Presumably, governments that are going to be spending, to meet the one-per-school-age-child target, billions of dollars on just the OLPC hardware and that have already had students, teachers, and education ministry staff involved in demonstration projects are going to roll the laptop out with something a little more sophisticated that dumping them on students with no instruction; the whole idea has been to integrate them into the curriculum. Of course, if you know of specific countries that have announced the intention to purchase the laptops and just dump them on school age children, please, feel free to point to them.
In what is, IIRC, the largest launch country, Brazil, median income for black women (the worst off racial/gender mix) is $156/month. (source)
Heck, even Rwanda (which is one of the poorest nations that may get it early, through Libya purchasing it for them) has an average per capita annual income of $206 (source), over an order of magnitude higher than you suggested for "most" OLPC recipients.
There is no such thing as individual "eligibility" for the laptops, so the question is incoherent. Yes, the US Department of Education is as free as any other national education ministry to purchase the laptops for distribution on a one-per-child basis, though of course they aren't the principal target market and the OLPC feature set is designed around use in a very different environment than one of the most developed nations in the world.
In many situations, higher-level languages reduce the cost of "good programming practices" by reducing the number of specific concerns that need to be addressed in a program designed for a given task, compared to C or C++.
They are not intended as a substitute for good practice, nor has anyone that I have seen suggest them as a substitute.
Eliminating the possibility of an entire class of bugs reduces the number of things that need to be remembered in putting the code together (and checked once it is coded), and therefore reduce the number of bugs (and exploitable vulnerabilities) that get through to end user code.
There are two arguments here:
The first is that, independently of the existence of any other standards covering the same subject matter, OOXML is a poorly described, non-implementable, and otherwise bad proposed standard, and should be rejected.
The second is that that the existence of one standard covering a topic makes additional standards covering the topic less valuable, potentially redundant, and in some cases contradictory to the purpose of standardization, particularly when adopted by the same standards body.
Debate over the second seems to only make sense in a context where it is assumed or concluded that OOXML would be a desirable standard on its own in the first place.
Tivo doesn't remove your right to modify the software now. The source is available. You can modify it. What they don't do is give you a system that will run modified software, nor an easy way to produce one. There are a few ways the FSF could have addressed that, presuming that concern is important (to me, it seems somewhat out-of-scope of the focus of the pre-v3 GPL, but it is a concern). One of them would have been to require equivalent openness of hardware: hardware that bundles GPL software would be required to have open specifications with rights that enable the recipient of the bundled hardware/software to "rebuild" their own implementation of the hardware (either a faithful one or a modified copy) and run whatever software they want on it. This would have allowed users freedom to do what they want, as well freedom to select hardware/software-bundle features.
Another approach, and the one the FSF chose—but only for consumers, because of outcry that business users needed the security of locked-down hardware/software bundles—was to have prohibit locked-down hardware. This requires it to be cheaper for users to run modified software than it might be under the other model, since they won't have to "rebuild" hardware, at the expense of the freedom to choose features of a hardware/software bundle they might prefer.
Or, looked at a different but equally accurate way, it is telling them (but not people who sell hardware with GPLv3 software that exclusively targets "business" users), "if you're giving our software to your users, you must not give them the option of hardware that provides an absolute guarantee against changes". IOW, it limits customer freedom to choose hardware features that they might be interested in.
Sure, hackers and tinkerers who are pretty certain that they'll be the most technically proficient users with physical access to their home hardware probably don't want that feature. Technophobic parents whose children are (or who perceive their children to be) far more technically proficient, OTOH, might prefer that feature. However, the FSF has declared that while businesses might have a need for hardware with that feature that makes it important to keep that choice available, consumers can't have that feature if they want OSS. In fact, it so important to keep that feature away from consumers that products that are "dual use" (marketed both to businesses and consumers) can't have the feature, only "business-only" products.
And if your business-only product is designed to be incorporated into a dwelling (say, you make a digital utility meter), well, then its a consumer product, and you can't have that feature, either.
If Microsoft, like many other online service providers, advertises or solicits business via the mail (certainly, they've done that for MSN, though I don't know if they have for Hotmail per se), it is governed by the same law that governs anyone else making such solicitations (not the USPS, but other postal service users).
OTOH, any online fraudulent solicitations by Microsoft would be more likely to be wire fraud, but Microsoft may be insulated from such charges from "free" users since Microsoft, while it uses them to get money from advertisers who hope to target them, does not get money or property from the users directly.
On the third hand, depending on how they market to advertisers, they may be guilty of fraud (regular, wire, mail, or all three) if they've misrepresented to them the kind of service their advertising will be associated with, since that is quite arguably a material misrepresentation directly to induce the advertiser to give money or property to Microsoft.
Since when do Iraq or the Government of Iraq have either peace or stability to threaten?
Wouldn't a determination that someone was threatening something that manifestly does not exist be, necessarily, unreasonable?
Or, rather, if executive branch officials without outside review decide that your demonstration suggests some potential future inclination toward violence.
How is it not a deprivation of property without due process?
Because the (approximate, though unfailingly somewhat biased) randomness of a die relies on uncertainties in physical conditions in the roll, whereas computers are designed to work predictably, to produce the same output to a given input. Even if you built a physical die roller into a computer system, I suspect it would be a lot of work to make it is good a random number generator as a human repeatedly rolling the same die.
Seems to me, Linux adoption has grown over the same time. So, if anything, the argument carries less weight now.
Unfortunately, the Linux (and open source more generally) community is, as a broad generality, much better at creating functioning software (particularly the parts "behind" the UI) than marketing or, to a lesser extent, making well-polished chrome; whereas the Windows "Community" (i.e., Microsoft) is much better at marketing and making chrome than building solid software internals.
It seems to me that "Loose" sounds like "Loose"—if you take most words in English that end with "-oose" as a guide, e.g., "moose", "goose", or "caboose".
OTOH, "lose" sounds like "Loose"—if you take "choose" as a model of how the "-oose" should sound.
I think the better statement is that "English spelling is only loosely connected to pronunciation".
Whether H. sapiens and H. neanderthalensis interbred to result in H. sapiens sapiens or whether, OTOH, H. sapiens sapiens evolved from H. sapiens alone before, after, or while the latter displaced H. neanderthalensis in Europe is somewhat irrelevant to the question of whether H. sapiens had a single origin in Africa.
It might be relevant to the question of whether H. sapiens sapiens had a single origin, which would be a different question. (A subspecies without a single origin would seem to be less surprising; a species with multiple origins would be, AFAIK, rather unusual.)
So?
Sure you can, because there aren't two different "ways" at issue for the same thing.
Sure, die-hard OSS advocates think that all software should be available under an OSS licensing model, both free and gratis.
That does not mean that they think that enterprises should just use the software and expect someone else to develope it to their needs and rely on free community support. Many OSS advocates are completely behind the idea of enterprises paying for support and development of OSS software (in fact, that's how many OSS advocates manage to eat) to meet their needs rather than paying for support and licensing of commercial software, selling as an advantage that, should their support vendor collapse for one reason or another, it will be easier for the enterprise to spend the same money to get the support for the software from another vendor or to develop support capacity in house than with commercial software where IP licensing concerns may make that more difficult to arrange without changing software.
Selling OSS as the ideal licensing approach doesn't mean that OSS is a free lunch for business users in terms of support. What it does mean is that the vendor choosing to go a different direction or being purchased by a competitor won't ever make support legally impossible. But nothing takes out the requirement that, for support to happen, there has to be someone willing to do the support, and that money will often be an important motivating factor in making that happen.
I don't know, do you think people should be polite to you or not?
No, this is completely true, and nothing you say "supporting" your argument that this is "wrong" even contradicts it, much less proves it wrong.
This is true, of course, but doesn't contradict the idea that you can take the code, modify it, and use the modified software on one's own hardware.
It also prevents people who are not the owner who get access to the device from running their choice of software on the Tivo DVR device. As the authors of the GPLv3 apparently recognized, the freedom to choose systems with this feature is important to some users. However, unfortunately, the GPLv3 authors decided that the freedom to include such a feature must be denied to consumers but preserved for business, with the preference for the former so strong that if a product has both consumer and business use, denying the freedom for consumers to select a product with such a feature overrides preserving the option for business.
That's possibly true, though the combination of the vague definition of "user product" in section 6 and the various terms that push all questionable or mixed cases into that category may create enough of a legal risk to prevent the use of such software in voting machines.
The GPLv3, whether one approves of it or not, is a sharp departure in that it incorporates restrictions on product features, which the GPLv2 did not, and that it has terms (those same restrictions) which substantially vary in effect based on who a product is marketed to and what is marketed for, whereas the GPLv2 had the same substantial effect no matter who a product was targetted at or how it was to be used.
Unlike the GPLv2, the GPLv3 is based, it appears from its substantive terms, not on the idea of a single, common set of "software freedoms", but on the idea that the FSF knows best which users ought to have which freedoms.
The FSF may or may not be analogous to a "church" with a "faith". But, AFAIK, Linus isn't a member of the FSF, and has never claimed to adhere to its "faith".
The GPLv2 isn't a church. Its a license. It has concrete terms. People (at least, some of them) decide to use it, or not, based on how well they determine those terms are aligned with their desires for their software. Linus, I expect, uses the GPLv2 for that reason, not out of quasi-religious faith in the cult of RMS.
I think Linus believes in freedom. I don't think he agrees with the FSF's understanding of what freedom is and how it should be defended as expressed in the GPLv3, though. Heck, I think that Linus's differences with the FSF on those points predate the GPLv3 discussion by quite a bit, though the GPLv2 was something he could live with though he understood some of its terms differently from the FSF.