Kernel 2.6.x is current Kernel 2.4.x is still supported Kernel 2.2.x is still supported Is that far enough back?
Clearly not for these guys - the last NT release was NT4 in 1996.
They are complaining at being forced to upgrade from that.
Comparison would be kernel 2.0 (or earlier if they are using earlier NT).
Good luck finding a Linux distribution vendor still supporting stuff this old.
Example: On the desktop, Redhat has already EOLed up to RHL9, which was 2002 or 2003 I think, and even with RHEL, support policy is 7 years from release. NT4 is ten years from release now.
Of course the great thing about FOSS is that you can support it yourself, so if it happens to be worth it you can hire your own kernel hackers (who are prepared to work on decade old versions).
For some large organisations, it might be worth it, but most people will go for upgrade after looking at the cost/risk.
Your analysis is not far wrong, but there is no need to involve copyright in this situation.
You buy a subsidised product based on paying for it over an N-month contract, and you either stick to that contract or you don't.
If you break the contract, they should sue you for that. Period.
The mobile phone market took off in the EU long before the EUCD broguht anything like the DMCA, and the "lock-down" is completely ineffective even now.
That the US mobile market is behind the EU is nothing to do with copyright - the US generally lags the richer EU nations on broadband penetration too, what is the copyright issue there ?
KJV translation also says: "in the covert of the reed" - which clearly does mean hidden by the reeds, not hiding behind a small hill whilst up to the ankles in reeds.
Your prehistoric = everything bigger argument is also fundamentally flawed:
1. it waves its tail like a cedar 2. cedars are like really big trees, therefore must have been a massive sauropod 3. [ but what about hiding in the reeds ] 4. oh but reeds were much much bigger back then - like everything else was
No massive holes in that argument, perfectly consistent. Not.
You missed the bit (conveniently ignored by those who argue it's a sauropod) about "in the cover of the reeds" or "hidden by the reeds".
Clearly the animal is small enough to be hidden in reeds in a river (reeds grow in shallow water, so you can't hide most of the dinosaur in the water either).
Exceptionally large it is not - if this is a literal description.
Doesn't sound like metaphor or a synonym either, it's a plain description of an animal small enough to hide in reeds. Might be a hippo, or an elephant, certainly isn't a Sauropod.
How about the ones (eg. Origen) that the later church fathers ordered burnt in about 500something AD.
Oh, slight problem there.
Just read something general about the early church history and you can at least verify that christians have been fighting, torturing and killing each other over this sort of belief for millenia.
[ Oops, sorry, must clarify: Christians don't kill/etc. each other, Christians kill heretics. You don't need to get into beliefs to understand who are the heretics, it's really simple: the hertic is the guy on the rack, the christian is the one holding the hot iron. ]
If you're from somewhere (like the UK) where driving licences don't have (or didn't used to have - ok, so they do now) photos, then it gets a bit tricky.
Passport was the only photo ID I had when I last went to the US.
Americans seem to view passports as more precious than we do on this side of the pond - probably because they can go thousands of miles without needing one - so they don't expect people to be carrying them.
This seemed to work with most door staff - explain you are english (not having a US accent is probably critical here) - explain that us wierd english don't have photos on ID other than passports - and obviously I didn't expect to have to carry my passport around from bar to bar all night
Wierd experience though - the last time I'd been asked my age (not asked for id) in a pub I was 19, and the question threw me _then_ because I hadn't been asked for a couple of years prior to that. Then, over a decade later (I was over 30) I go the the US and I'm being asked to prove my age at every bar. Now, I know the legal drinking age was 21 vs. 18 in the UK, but even so, could I really look over ten years younger than I was ? Freaky.
So, for example, you first ensure you understand your users' training needs by running a pilot project, trying different levels of training and testing and getting user feedback.
Then you find that training costs on Linux are higher because fewer people are familiar with it and they don't like change.
So:
1. you "do whatever it was that..." - by training your users up on Linux first so now Windows is not cheaper.
2. Now you save money by deploying Linux instead.
3. Get big bonus for saving lots of money
4. Get audited and get fired over the "miscellaneous Linux training" line item that cost three times what you saved on the project, that you failed to bury deep enough in the books.
Big organisations have lots of inertia. Unseating the incumbent is hard. The no-change option is low-risk so you have to be a lot better and a lot cheaper.
Linux on the server got in through the back door in many cases - interested sysadmins simply installed it, used it, proved it (eliminating the risk argument) and then made it official afterwards. Much, much, harder to do that on the desktop.
The next big window of opportunity for Linux on the Desktop is probably in 12-18months when large users really start looking at Vista / Office2007 (well, they'll be looking at it right now, but not with a view to roll out tomorrow), and earlier Office versions start to move out of support (I think win2k already has?). Re-training and risk are then back in there on the microsoft side too, makeing the playing field a bit more level.
Plenty (probably most) religious people will enter the churches/temples of other religions for things like weddings. No reason why aetheists should be different, nor reason to assume they are simply afraid.
I've been to Jewish, Muslim, Hindu, Protestant, and Catholic weddings as well as secular, and I've seen people of various religions at those weddings.
Sure, there are Protestants who wouldn't attend a Catholic wedding (Ian Paisley comes to mind...) and Christians who wouldn't enter a mosque etc. - but those tend to be the extremists not the norm (and the GP didn't say "all aetheists", it said "most").
There are plenty of good reasons for going to the bar rather than the ceremony - reasons that don't involve religion.
There are plenty of four-way stops (or at least give-ways / yields) in the uk - usually on very minor roads in built-up areas as part of traffic calming. Often the whole intersection area is raised like a large speed-bump.
The fact that no one knows who has priority does seem to slow everything down.
Not sure what happens if there is an accident though - presumably everyone is at fault:-)
Why is no-one funding you? [...] and if I don't get funding here in America I'll give my patents to China for free and you wouldn't want that.
Not what I heard. He said the research is out there now for anyone to pick up, he'd like it to be America first, but that's not up to him to decide now. America stopped funding, so he's out of embargo, so he's published.
Why (how) on earth would he "give" patents to China ? You think China honours US patents ? LOL.
The patents are going to be gone soon anyhow - one at least is 89, I think the other is 92, and isn't it 17yrs for US patents ?
This project, probably Focus Fusion if I read the tealeaves right without even seeing the videos
It's not - well maybe it is if you think alpha-particle beam rather than proton, however it still may not be as you know it.
It is worth watching, in part because it _isn't_ Farnsworth Fusor all over again (and he goes into why Farnsworth won't work as well as damning Tokamak).
According to the video, this guy's R&D team have been on this for over a decade with a US Navy grant + publication embargo - but their grant was killed so now they can talk.
They started at Farnsworth, figured that grid-impact losses made it a non-starter and took away the grid, using a new take on magnetic confinement (polyhedral / psuedo-spherical) of injected high energy electrons to create the potential well.
They claim to have got to several orders of magnitude better than Farnsworth devices, and only a couple short of breakeven. Very interesting. Might still be snake oil, but Farnsworth isn't snake oil (just never going to get to breakeven) and this does sound like a plausible research avenue for a way to fix the losses with Farnsworth devices.
Video is very amusing in places - particularly some of the bits about Tokamak / Jet / ITER. To paraphrase one quote (because I can't be bothered to seek through the whole video again): "we know fusion works, go outside look up in the night sky, billions of working fusion reactors... none of them toroidal". Sweet - I'm going to remember that one.
4. Novell is STILL trying to sell fricking Netware. Doesn't seem to notice that nobody fricking gives two turds about Netware anymore.
Sadly in the big corporate world that just isn't true, I wish it was (would make life a lot simpler). I get asked for Netware / eDirectory / Groupwise (you missed that one) integration all the time. The people asking aren't asking for nostalgia - they are running networks with 10k+ desktops on those products right now.
Where did I say that? You put that in quotation marks as if it were a word by word citation
It was - but I see now how I haven't parsed the clauses in the way you intended.
However my point still stands - for patent licences it is irrelevant who developed or holds the copyright on the software. Even "use" of the software may infringe let alone distribution.
What other non-MS patents are you referring to? Do you have specific examples
Kodak (proven in court). Eolas (alleged). Unisys (gif/compression, some have now expired but "Unisys Corporation holds and has patents pending on a number of improvements on the inventions claimed in the above-expired patents."). Forgent (Jpeg - may be declared invalid now?). Hibernate is also under patent threat - not sure if that is in Suns Java dist. yet, but it is so widely used that surely it will get added now (and is (L)GPL).
Sun is under at least one patent licence agreement with Kodak, that settled that case - but we don't know the terms, are they GPL compatible or not, maybe they only cover Sun and anyone Sun ships to ?
we do not believe there is a need for or basis for the type of relationship defined in the Microsoft-Novell announcement
Must have more money than they know what to do with then.
Remember what type of relationship this is: Novell just got $300M plus from MS in total for this mutual covenant-not-to-sue (and some sales "cooperation" / coupons).
Why on earth would anyone turn down that ?
Let's face it, RedHat are saying they aren't interested in order to get Kudos, because they know Ballmer is not interested anyway. He needs more deals like this like he needs a chair in the head. He's spouted a bunch of FUD to convince _his shareholders_ that he's done a brilliant deal, when in reality he's had to make a massive pay out. So, obviously, he is also going to _say_ that he wants to make more such "brilliant deals" - in reality it's the last thing he wants. Sure, MS may have billions in cash, but do you really think they want to go handing it out 300M at a time to dozens of Linux companies ?
If I really thought Ballmer was serious about doing the same deal with other Linux distributions I'd be setting up a Linux distribution tomorrow (and hey, $50M will do me Mr B, quick easy deal, I'm not greedy, I'll even throw in a joint press conf.).
Now you seem to be even more confused - "Novell is distributing software developed by MS" - huh ? This is about patents, not copyright. It matters not at all who developed the software - patent liability doesn't care. And what MS software is Novell shipping ???
Note that I have never claimed that the deals are exactly the same - they clearly aren't - but what I am interested in is what is the difference that makes one deal ok by the GPL and the other an (alleged) violation.
The GPL argument, as I understand it, is that any patent licence you have that covers GPL code has to be freely distributable (to everyone downstream) - or you cannot distribute the code. Liberty or death. Note that this is about licences you have to 3rd party patents - your patents are automatically licenced by distribution under GPL, so how many patents Novel & Sun have is not relevant - it's the ones they licence from elsewhere (MS) that are important for the liberty-or-death clause.
Agreed or not ?
Now the deals:
Both deals involve paying royalties to MS in respect of MS patents in return for some sort of covenant-not-to-sue (which is frequently argued to be identical to a patent licence).
In both cases, the licenced MS patents may or may not cover the GPL code and may or may not be valid etc. etc. - but we don't know. Although, in fact Novell says that there are no patent issues with the code they ship, which may cover them wrt. GPL. Ballmer says the MS patents do cover Linux, but then he's been saying that for years and he's full of it - why should we believe him ? Sun, on the other hand, hasn't said that Java is clean - probably because they'd be lying if they did, as Java is known to violate third party patents.
But, leaving the above aside because we don't know which patents / code (if any) is at issue in either case (so, no difference there), the argument frequently made against Novell is that the covenant/licence itself is the issue, because why would they need it, why would they pay royalties (back) to MS, if there was no violation ? Now, I'm not entirely convinced by that argument, but assuming it applies, then it also applies to Sun, since they are paying MS for a patent licence/covenant too.
So, at this point we conclude that either by the GPL liberty-or-death argument, the patent covenant/licences must be freely distributable downstream to meet the GPL. [Novell-MS's is not, Sun's might be, I don't have a copy of it to tell].
Or they don't have to be distributable because in some way they don't cover the GPL code.
Either way it must be the same for both cases, since the evidence of a patent licence exisitng and covering GPL code is the same in each case (Novell & MS statements on applicability appear to disagree, so discount both, Sun has no statement).
[I'm assuming here that we see through Novell-MS's wierd "party pays the royalties, party's customers get the licence" scheme as being really just a traditional cross licence like the GPL was designed to account for].
And finally, you still haven't responded on the issue of the other non-MS patents that Sun licences for Java - are those patent licences GPL-compatible (ie. freely redistributable to everyone) or not ?
It's NOT a violation of the notion of free trade to ban or restrict items from other countries that are ALREADY banned or restricted domestically.
Indeed it isn't - nor would it be a WTO violation. The problem, as I understand the WTO argument, is that the US _does_ allow online gambling but only for US companies. That breaks WTO rules just like if the US banned the sale of non-US oranges - the US is still free to ban the sale of oranges altogether.
Note also that WTO doesn't affect sovereignty - it's a mutual agreement thing, and it doesn't have any enforcement as such. The US is actually free to break WTO rules, however if it does then the vicitms are entitled to charge the US back with levys etc. (withing the rules). The US could then retaliate further if it wanted - but things tend to stop at that point because global trade war is not in any of the WTO players' interest, so they work it out.
The interesting thing here is that Antigua is so small it can't recover its damages from the US in the usual fashion, so it is asking for the novel relief of being granted the right to copy US-produced IP without paying the usual royalties. Since global enforcement of copyright relies on similar mutual agreement to WTO (might even come under WTO?), this might even work.
Not that the "SUN gpl-ed JAVA, is friendly to GPL v3 " neglects to mention that Sun is party to a mutual covenant-not-to-sue patent agreement with regards to Java, and it is with......Microsoft.
Just like Novell. Oh. So what is the GPLv3 going to say ?
"if you have a covenant-not-to-sue deal that is really a patent-licence that coevers this software, then it must be freely sublicenceable to all - if you are Novell or MS, if you are Sun you can ignore this bit"
[that's before we get onto the other patents that Sun has had to pay to licence for Java - are those licences freely availabel for all as the GPL requires... or not ?]
It's more than that. Look at the other parts of the deal - total is $300M+.
MS bought a lot of "coupons" for Suse. Good sell to shareholders (only $100M for patents and we get $40M back from them) - just leave out the "we bought some paper for $240M and we paid $300M+ up front and we'll only get the $40M back as a % of Novell sales when they happen".
It's a payoff for Novell (for Novell's IP in MS products) - exactly like the one for Sun a couple of years ago.
ie. 2/3 is the maximum (in fibre) - for when you are trying for fast as possible. pick a different material and you can get several orders of magnitude slower, if that is what you desire.
Slashdot Mirror
If you know so much then tell me where my copy of the Hitchhiker's Trilogy has gotten off to.
It's run off with your biros.
Doesn't the tooth comb make, er..., scratches ?
Kernel 2.6.x is current
Kernel 2.4.x is still supported
Kernel 2.2.x is still supported
Is that far enough back?
Clearly not for these guys - the last NT release was NT4 in 1996.
They are complaining at being forced to upgrade from that.
Comparison would be kernel 2.0 (or earlier if they are using earlier NT).
Good luck finding a Linux distribution vendor still supporting stuff this old.
Example: On the desktop, Redhat has already EOLed up to RHL9, which was 2002 or 2003 I think, and even with RHEL, support policy is 7 years from release. NT4 is ten years from release now.
Of course the great thing about FOSS is that you can support it yourself, so if it happens to be worth it you can hire your own kernel hackers (who are prepared to work on decade old versions).
For some large organisations, it might be worth it, but most people will go for upgrade after looking at the cost/risk.
Your analysis is not far wrong, but there is no need to involve copyright in this situation.
You buy a subsidised product based on paying for it over an N-month contract, and you either stick to that contract or you don't.
If you break the contract, they should sue you for that. Period.
The mobile phone market took off in the EU long before the EUCD broguht anything like the DMCA, and the "lock-down" is completely ineffective even now.
That the US mobile market is behind the EU is nothing to do with copyright - the US generally lags the richer EU nations on broadband penetration too, what is the copyright issue there ?
It clearly says UNDER a lotus plant.
Which still wouldn't fit a sauropod.
KJV translation also says: "in the covert of the reed" - which clearly does mean hidden by the reeds, not hiding behind a small hill whilst up to the ankles in reeds.
Your prehistoric = everything bigger argument is also fundamentally flawed:
1. it waves its tail like a cedar
2. cedars are like really big trees, therefore must have been a massive sauropod
3. [ but what about hiding in the reeds ]
4. oh but reeds were much much bigger back then - like everything else was
No massive holes in that argument, perfectly consistent. Not.
You missed the bit (conveniently ignored by those who argue it's a sauropod) about "in the cover of the reeds" or "hidden by the reeds".
Clearly the animal is small enough to be hidden in reeds in a river (reeds grow in shallow water, so you can't hide most of the dinosaur in the water either).
Exceptionally large it is not - if this is a literal description.
"hidden among the reeds in the marsh."
Doesn't sound like metaphor or a synonym either, it's a plain description of an animal small enough to hide in reeds. Might be a hippo, or an elephant, certainly isn't a Sauropod.
21 Under the lotus plants he lies,
hidden among the reeds in the marsh.
And exactly what kind of reeds are you proposing to hide a Sauropod in ?
Whatever it is, it is small enought to "lie under a lotus plant" (a few metres tall at most) and be "hidden by reeds" (a couple of metres tall).
Doesn't sound like a sauropod to me. More like a hippo.
How about the ones (eg. Origen) that the later church fathers ordered burnt in about 500something AD.
Oh, slight problem there.
Just read something general about the early church history and you can at least verify that christians have been fighting, torturing and killing each other over this sort of belief for millenia.
[ Oops, sorry, must clarify: Christians don't kill/etc. each other, Christians kill heretics. You don't need to get into beliefs to understand who are the heretics, it's really simple: the hertic is the guy on the rack, the christian is the one holding the hot iron. ]
If you're from somewhere (like the UK) where driving licences don't have (or didn't used to have - ok, so they do now) photos, then it gets a bit tricky.
Passport was the only photo ID I had when I last went to the US.
Americans seem to view passports as more precious than we do on this side of the pond - probably because they can go thousands of miles without needing one - so they don't expect people to be carrying them.
This seemed to work with most door staff
- explain you are english (not having a US accent is probably critical here)
- explain that us wierd english don't have photos on ID other than passports
- and obviously I didn't expect to have to carry my passport around from bar to bar all night
Wierd experience though - the last time I'd been asked my age (not asked for id) in a pub I was 19, and the question threw me _then_ because I hadn't been asked for a couple of years prior to that. Then, over a decade later (I was over 30) I go the the US and I'm being asked to prove my age at every bar. Now, I know the legal drinking age was 21 vs. 18 in the UK, but even so, could I really look over ten years younger than I was ? Freaky.
So, for example, you first ensure you understand your users' training needs by running a pilot project, trying different levels of training and testing and getting user feedback.
Then you find that training costs on Linux are higher because fewer people are familiar with it and they don't like change.
So:
1. you "do whatever it was that..." - by training your users up on Linux first so now Windows is not cheaper.
2. Now you save money by deploying Linux instead.
3. Get big bonus for saving lots of money
4. Get audited and get fired over the "miscellaneous Linux training" line item that cost three times what you saved on the project, that you failed to bury deep enough in the books.
Big organisations have lots of inertia. Unseating the incumbent is hard. The no-change option is low-risk so you have to be a lot better and a lot cheaper.
Linux on the server got in through the back door in many cases - interested sysadmins simply installed it, used it, proved it (eliminating the risk argument) and then made it official afterwards. Much, much, harder to do that on the desktop.
The next big window of opportunity for Linux on the Desktop is probably in 12-18months when large users really start looking at Vista / Office2007 (well, they'll be looking at it right now, but not with a view to roll out tomorrow), and earlier Office versions start to move out of support (I think win2k already has?). Re-training and risk are then back in there on the microsoft side too, makeing the playing field a bit more level.
Plenty (probably most) religious people will enter the churches/temples of other religions for things like weddings. No reason why aetheists should be different, nor reason to assume they are simply afraid.
I've been to Jewish, Muslim, Hindu, Protestant, and Catholic weddings as well as secular, and I've seen people of various religions at those weddings.
Sure, there are Protestants who wouldn't attend a Catholic wedding (Ian Paisley comes to mind...) and Christians who wouldn't enter a mosque etc. - but those tend to be the extremists not the norm (and the GP didn't say "all aetheists", it said "most").
There are plenty of good reasons for going to the bar rather than the ceremony - reasons that don't involve religion.
There are plenty of four-way stops (or at least give-ways / yields) in the uk - usually on very minor roads in built-up areas as part of traffic calming. Often the whole intersection area is raised like a large speed-bump.
:-)
The fact that no one knows who has priority does seem to slow everything down.
Not sure what happens if there is an accident though - presumably everyone is at fault
No slight intended on the product - I wouldn't want to support BetaMax now, but that doesn't mean I think VHS is better.
Why is no-one funding you? [...] and if I don't get funding here in America I'll give my patents to China for free and you wouldn't want that.
Not what I heard. He said the research is out there now for anyone to pick up, he'd like it to be America first, but that's not up to him to decide now. America stopped funding, so he's out of embargo, so he's published.
Why (how) on earth would he "give" patents to China ? You think China honours US patents ? LOL.
The patents are going to be gone soon anyhow - one at least is 89, I think the other is 92, and isn't it 17yrs for US patents ?
This project, probably Focus Fusion if I read the tealeaves right without even seeing the videos
It's not - well maybe it is if you think alpha-particle beam rather than proton, however it still may not be as you know it.
It is worth watching, in part because it _isn't_ Farnsworth Fusor all over again (and he goes into why Farnsworth won't work as well as damning Tokamak).
According to the video, this guy's R&D team have been on this for over a decade with a US Navy grant + publication embargo - but their grant was killed so now they can talk.
They started at Farnsworth, figured that grid-impact losses made it a non-starter and took away the grid, using a new take on magnetic confinement (polyhedral / psuedo-spherical) of injected high energy electrons to create the potential well.
They claim to have got to several orders of magnitude better than Farnsworth devices, and only a couple short of breakeven. Very interesting. Might still be snake oil, but Farnsworth isn't snake oil (just never going to get to breakeven) and this does sound like a plausible research avenue for a way to fix the losses with Farnsworth devices.
Video is very amusing in places - particularly some of the bits about Tokamak / Jet / ITER. To paraphrase one quote (because I can't be bothered to seek through the whole video again): "we know fusion works, go outside look up in the night sky, billions of working fusion reactors... none of them toroidal". Sweet - I'm going to remember that one.
4. Novell is STILL trying to sell fricking Netware. Doesn't seem to notice that nobody fricking gives two turds about Netware anymore.
Sadly in the big corporate world that just isn't true, I wish it was (would make life a lot simpler). I get asked for Netware / eDirectory / Groupwise (you missed that one) integration all the time. The people asking aren't asking for nostalgia - they are running networks with 10k+ desktops on those products right now.
Where did I say that? You put that in quotation marks as if it were a word by word citation
It was - but I see now how I haven't parsed the clauses in the way you intended.
However my point still stands - for patent licences it is irrelevant who developed or holds the copyright on the software. Even "use" of the software may infringe let alone distribution.
What other non-MS patents are you referring to? Do you have specific examples
Kodak (proven in court). Eolas (alleged). Unisys (gif/compression, some have now expired but "Unisys Corporation holds and has patents pending on a number of improvements on the inventions claimed in the above-expired patents."). Forgent (Jpeg - may be declared invalid now?). Hibernate is also under patent threat - not sure if that is in Suns Java dist. yet, but it is so widely used that surely it will get added now (and is (L)GPL).
Sun is under at least one patent licence agreement with Kodak, that settled that case - but we don't know the terms, are they GPL compatible or not, maybe they only cover Sun and anyone Sun ships to ?
Must have more money than they know what to do with then.
Remember what type of relationship this is: Novell just got $300M plus from MS in total for this mutual covenant-not-to-sue (and some sales "cooperation" / coupons).
Why on earth would anyone turn down that ?
Let's face it, RedHat are saying they aren't interested in order to get Kudos, because they know Ballmer is not interested anyway. He needs more deals like this like he needs a chair in the head. He's spouted a bunch of FUD to convince _his shareholders_ that he's done a brilliant deal, when in reality he's had to make a massive pay out. So, obviously, he is also going to _say_ that he wants to make more such "brilliant deals" - in reality it's the last thing he wants. Sure, MS may have billions in cash, but do you really think they want to go handing it out 300M at a time to dozens of Linux companies ?
If I really thought Ballmer was serious about doing the same deal with other Linux distributions I'd be setting up a Linux distribution tomorrow (and hey, $50M will do me Mr B, quick easy deal, I'm not greedy, I'll even throw in a joint press conf.).
And I refuted your non-answers.
Now you seem to be even more confused - "Novell is distributing software developed by MS" - huh ? This is about patents, not copyright. It matters not at all who developed the software - patent liability doesn't care. And what MS software is Novell shipping ???
Note that I have never claimed that the deals are exactly the same - they clearly aren't - but what I am interested in is what is the difference that makes one deal ok by the GPL and the other an (alleged) violation.
The GPL argument, as I understand it, is that any patent licence you have that covers GPL code has to be freely distributable (to everyone downstream) - or you cannot distribute the code. Liberty or death. Note that this is about licences you have to 3rd party patents - your patents are automatically licenced by distribution under GPL, so how many patents Novel & Sun have is not relevant - it's the ones they licence from elsewhere (MS) that are important for the liberty-or-death clause.
Agreed or not ?
Now the deals:
Both deals involve paying royalties to MS in respect of MS patents in return for some sort of covenant-not-to-sue (which is frequently argued to be identical to a patent licence).
In both cases, the licenced MS patents may or may not cover the GPL code and may or may not be valid etc. etc. - but we don't know. Although, in fact Novell says that there are no patent issues with the code they ship, which may cover them wrt. GPL. Ballmer says the MS patents do cover Linux, but then he's been saying that for years and he's full of it - why should we believe him ? Sun, on the other hand, hasn't said that Java is clean - probably because they'd be lying if they did, as Java is known to violate third party patents.
But, leaving the above aside because we don't know which patents / code (if any) is at issue in either case (so, no difference there), the argument frequently made against Novell is that the covenant/licence itself is the issue, because why would they need it, why would they pay royalties (back) to MS, if there was no violation ? Now, I'm not entirely convinced by that argument, but assuming it applies, then it also applies to Sun, since they are paying MS for a patent licence/covenant too.
So, at this point we conclude that either by the GPL liberty-or-death argument, the patent covenant/licences must be freely distributable downstream to meet the GPL. [Novell-MS's is not, Sun's might be, I don't have a copy of it to tell].
Or they don't have to be distributable because in some way they don't cover the GPL code.
Either way it must be the same for both cases, since the evidence of a patent licence exisitng and covering GPL code is the same in each case (Novell & MS statements on applicability appear to disagree, so discount both, Sun has no statement).
[I'm assuming here that we see through Novell-MS's wierd "party pays the royalties, party's customers get the licence" scheme as being really just a traditional cross licence like the GPL was designed to account for].
And finally, you still haven't responded on the issue of the other non-MS patents that Sun licences for Java - are those patent licences GPL-compatible (ie. freely redistributable to everyone) or not ?
It's NOT a violation of the notion of free trade to ban or restrict items from other countries that are ALREADY banned or restricted domestically.
Indeed it isn't - nor would it be a WTO violation. The problem, as I understand the WTO argument, is that the US _does_ allow online gambling but only for US companies. That breaks WTO rules just like if the US banned the sale of non-US oranges - the US is still free to ban the sale of oranges altogether.
Note also that WTO doesn't affect sovereignty - it's a mutual agreement thing, and it doesn't have any enforcement as such. The US is actually free to break WTO rules, however if it does then the vicitms are entitled to charge the US back with levys etc. (withing the rules). The US could then retaliate further if it wanted - but things tend to stop at that point because global trade war is not in any of the WTO players' interest, so they work it out.
The interesting thing here is that Antigua is so small it can't recover its damages from the US in the usual fashion, so it is asking for the novel relief of being granted the right to copy US-produced IP without paying the usual royalties. Since global enforcement of copyright relies on similar mutual agreement to WTO (might even come under WTO?), this might even work.
Not that the "SUN gpl-ed JAVA, is friendly to GPL v3 " neglects to mention that Sun is party to a mutual covenant-not-to-sue patent agreement with regards to Java, and it is with... ...Microsoft.
Just like Novell. Oh. So what is the GPLv3 going to say ?
"if you have a covenant-not-to-sue deal that is really a patent-licence that coevers this software, then it must be freely sublicenceable to all - if you are Novell or MS, if you are Sun you can ignore this bit"
[that's before we get onto the other patents that Sun has had to pay to licence for Java - are those licences freely availabel for all as the GPL requires... or not ?]
It's more than that. Look at the other parts of the deal - total is $300M+.
MS bought a lot of "coupons" for Suse. Good sell to shareholders (only $100M for patents and we get $40M back from them) - just leave out the "we bought some paper for $240M and we paid $300M+ up front and we'll only get the $40M back as a % of Novell sales when they happen".
It's a payoff for Novell (for Novell's IP in MS products) - exactly like the one for Sun a couple of years ago.
point is (as you say) "as much as".
ie. 2/3 is the maximum (in fibre) - for when you are trying for fast as possible. pick a different material and you can get several orders of magnitude slower, if that is what you desire.