GPLv3 has a significantly more nasty viral nature than GPLv2, as the Anti-TiVo clause and the Anti-Patent clause, as well as the significant expansion of the term "conveying", makes the GPLv3 much more dangerous for a business to deal with/use.
Thus I would be concerned myself that GPLv3 will reduce the adoption of the open-source code, as it get farther and farther away from the BSD model of "do whatever you want".
I don't care for racing games: Motostorm is "Ohh, pretty, Ohh Shiney", but I wouldn't play more than an hour of it.
So yeah, its a "who cares" game and a feature reduction (because if the software compatibility was so superior, why aren't they running it on the 1.0 hardware when appropriate?).
Sony is phasing out the 100% backwards compatible model, replacing it with a model which STILL costs way too much at the original price, only gains a "who cares" game and 20 GB of hard drives space...
While at the same time, replaces the quality backwards compatibility with a incomplete software layer...
Now tell me why I shouldn't just get an XBox instead. The hardware may be unreliable, but Microsoft will at least fix it when it breaks.
This, as other posters have mentioned, is a way of GPL3-proofing CUPS, which is a key piece of the apple architecture and made possible because the copyrights for contributions were transfered to the developers.
Unlike Microsoft, Apple depends pretty heavily on GPL'ed code: CUPS, dev tools, and a lot of assorted *nix tools. Under GPL2 they were happy: release the source for the tools, any modifications they made, and be happy.
GPL3 is a nightmare: both the anti-Tivo clause and the anti-Patent clause represent huge and unacceptable changes to Apple. The anti-Tivo clause goes against the iPhone, iPod, MacTV, etc. And the anti-patent clause represents unilateral disarmament in the defensive-patent war, so even if you weren't going to enforce the patents, just have them for defense, GPL3 is a vulnerability.
So expect the following:
When possible, Apple will buy out any developers who own copyrights on GPL'ed code they depend on.
Otherwise, two things will happen: Apple will feature-freeze with a GPL2 version and fork, or simply replace the GPL'ed code completely.
Congratulations, RMS, I think this is what you actually intended. And it will work. Enjoy.
Microsoft will do NOTHING involving anything with GPLv3. THey already view the GPL as a dangerous virus and are quite particular about keeping that contamination out of their ecology.
Those "unexpiring" vouchers won't cover GPLv3 stuff, and even if it DID, it is highly unlikely a court would enforce the patent covenants.
So when Microsoft says it is unaffected by the GPLv3, that is perfectly correct, they will have NOTHING to do with it, and anything otherwise is wishful dreaming.
Microsoft has always viewed the GPL as a virus, and has made all attempts to avoid contact. Their paranoia on this front is legendary, and for good reason when you hear the GPL crowd react to actual or perceived violations of the GPL.
With the Support certificates, microsoft was deliberately having a competitor actually handle the support and touch the GPL code. Which is all fine and good under GPL2.
The GPL3 patent covenant is even more toxic, especially to a company like Microsoft which has a lot of patents. So they are simply saying "our certificates will not support anything on GPLv3".
In many ways, this is Microsoft's paranoid overreaction, as they are not by any means a contributor to the code, even if the certificats were valid for GPLv3, but it is an understandable conservative reaction.
Since Microsoft has never and WILL NEVER contribute or distribute GPLv3 code, yes, the statement is perfectly correct.
Assuming they just do full replace and junk, and pay full retail price, that is >2.5 MILLION failed X-Boxes in the next 2 years. Assuming each repair costs Microsoft only $200, they are budgeting for 5 MILLION failed x-boxen!
With only 11 million X-boxen shipped, that 33% failure rate is sounding like an UNDERCOUNT!
Look at the results of the Elevator 2010 competition:
They are getting within 20% of the theoretical limit of the fibers, and that is 50x less than where they need to be.
If you could make a carbon nanotube rope which is even just 1 inch long and which shows an appreciable amount of the raw nanotube strength, you can get your 10M in VC funding in, ohh, about 2 seconds flat.
You need materials which are NEARLY TWO ORDERS OF MAGNITUDE stronger than what we can build today...
And if you got just 50% stronger, let alone 5000% stronger, you would have a hugely viable company just making cables and fibers.
IBM etc all predicate their work on a reasonable roadmap of the future. Any roadmap which says "And then a miracle occures, and we are able to improve things by 50x when previously we've only been able to improve things by 1.01x" is just not a viable roadmap.
If you use them a lot, the ink costs a fortune per page, several times what a laser printer costs per page, so that a more expensive laser printer quickly becomes cheaper to own.
But if you don't use them a lot, the ink evaporates and the inkjet clogs up and stops working, forcing you to buy even more cartridges, so that a more expensive laser printer quickly becomes cheaper to own.
OK, he doesn't seem QUITE so loony. But I remember seeing some nice proofs how EPR doesn't violate causality, because it can't transmit classical information.
But I don't remember how those proofs work, because quantum gives me a headache.
If relativity is correct (and even possibly if it isn't), backwards-in-time communication really REALLY F@#)(*s up causality. Heck, Faster Than Light (FTL) communication at all F@#)(*s up causality.
The EPR paradox, in particular, can't actually communicate information, as you can't actually tell which of the two sides measured it first (or at all), preventing FTL communication and saving causality. (Of course, simultinaity doesn't make sense either, thanks to relativity, but causality does).
Long LONG before you can build a space elevator you need tether materials which are several orders of magnitude stronger than what we can build today...
If you could even get 1/100th of the way there on materials, you would have a great company selling fibers for military and industrial applications.
A compromised card which is revoked but not used by an attacker COSTS money! Big money.
Its only if a compromised card is USED by an attacker that there is a problem. But since cards get stolen as well, they have heavy misuse detection to catch this, and if they let a few slip through, they aren't the ones holding the bill anyway, as it usually ends up being chargebacked to the merchant who accepted the bogus card.
Slashdot Mirror
And here I thought it was Sunday the 15th.
./ editors. The dupes are getting really annoying.
Come on
GPLv3 has a significantly more nasty viral nature than GPLv2, as the Anti-TiVo clause and the Anti-Patent clause, as well as the significant expansion of the term "conveying", makes the GPLv3 much more dangerous for a business to deal with/use.
Thus I would be concerned myself that GPLv3 will reduce the adoption of the open-source code, as it get farther and farther away from the BSD model of "do whatever you want".
I don't care for racing games: Motostorm is "Ohh, pretty, Ohh Shiney", but I wouldn't play more than an hour of it.
So yeah, its a "who cares" game and a feature reduction (because if the software compatibility was so superior, why aren't they running it on the 1.0 hardware when appropriate?).
Sony is phasing out the 100% backwards compatible model, replacing it with a model which STILL costs way too much at the original price, only gains a "who cares" game and 20 GB of hard drives space...
While at the same time, replaces the quality backwards compatibility with a incomplete software layer...
Now tell me why I shouldn't just get an XBox instead. The hardware may be unreliable, but Microsoft will at least fix it when it breaks.
This, as other posters have mentioned, is a way of GPL3-proofing CUPS, which is a key piece of the apple architecture and made possible because the copyrights for contributions were transfered to the developers.
Unlike Microsoft, Apple depends pretty heavily on GPL'ed code: CUPS, dev tools, and a lot of assorted *nix tools. Under GPL2 they were happy: release the source for the tools, any modifications they made, and be happy.
GPL3 is a nightmare: both the anti-Tivo clause and the anti-Patent clause represent huge and unacceptable changes to Apple. The anti-Tivo clause goes against the iPhone, iPod, MacTV, etc. And the anti-patent clause represents unilateral disarmament in the defensive-patent war, so even if you weren't going to enforce the patents, just have them for defense, GPL3 is a vulnerability.
So expect the following:
When possible, Apple will buy out any developers who own copyrights on GPL'ed code they depend on.
Otherwise, two things will happen: Apple will feature-freeze with a GPL2 version and fork, or simply replace the GPL'ed code completely.
Congratulations, RMS, I think this is what you actually intended. And it will work. Enjoy.
Microsoft will do NOTHING involving anything with GPLv3. THey already view the GPL as a dangerous virus and are quite particular about keeping that contamination out of their ecology.
Those "unexpiring" vouchers won't cover GPLv3 stuff, and even if it DID, it is highly unlikely a court would enforce the patent covenants.
So when Microsoft says it is unaffected by the GPLv3, that is perfectly correct, they will have NOTHING to do with it, and anything otherwise is wishful dreaming.
2+ years ago I had this happen to me. It is an easy way to validate old card #s that a hacker gets.
Microsoft has always viewed the GPL as a virus, and has made all attempts to avoid contact. Their paranoia on this front is legendary, and for good reason when you hear the GPL crowd react to actual or perceived violations of the GPL.
With the Support certificates, microsoft was deliberately having a competitor actually handle the support and touch the GPL code. Which is all fine and good under GPL2.
The GPL3 patent covenant is even more toxic, especially to a company like Microsoft which has a lot of patents. So they are simply saying "our certificates will not support anything on GPLv3".
In many ways, this is Microsoft's paranoid overreaction, as they are not by any means a contributor to the code, even if the certificats were valid for GPLv3, but it is an understandable conservative reaction.
Since Microsoft has never and WILL NEVER contribute or distribute GPLv3 code, yes, the statement is perfectly correct.
Assuming they just do full replace and junk, and pay full retail price, that is >2.5 MILLION failed X-Boxes in the next 2 years. Assuming each repair costs Microsoft only $200, they are budgeting for 5 MILLION failed x-boxen!
With only 11 million X-boxen shipped, that 33% failure rate is sounding like an UNDERCOUNT!
Vendor benchmarks are always considered untrustworthy, so I don't see what the big deal is.
Look at the results of the Elevator 2010 competition:
They are getting within 20% of the theoretical limit of the fibers, and that is 50x less than where they need to be.
If you could make a carbon nanotube rope which is even just 1 inch long and which shows an appreciable amount of the raw nanotube strength, you can get your 10M in VC funding in, ohh, about 2 seconds flat.
You need materials which are NEARLY TWO ORDERS OF MAGNITUDE stronger than what we can build today...
And if you got just 50% stronger, let alone 5000% stronger, you would have a hugely viable company just making cables and fibers.
IBM etc all predicate their work on a reasonable roadmap of the future. Any roadmap which says "And then a miracle occures, and we are able to improve things by 50x when previously we've only been able to improve things by 1.01x" is just not a viable roadmap.
If you use them a lot, the ink costs a fortune per page, several times what a laser printer costs per page, so that a more expensive laser printer quickly becomes cheaper to own.
But if you don't use them a lot, the ink evaporates and the inkjet clogs up and stops working, forcing you to buy even more cartridges, so that a more expensive laser printer quickly becomes cheaper to own.
Buy a laser printer based all-in-one.
Can't you just make a binary blob kernel module? I know the GPL zealots hate the idea, but wouldn't this also get around the liscencing problems?
OK, he doesn't seem QUITE so loony. But I remember seeing some nice proofs how EPR doesn't violate causality, because it can't transmit classical information.
But I don't remember how those proofs work, because quantum gives me a headache.
If relativity is correct (and even possibly if it isn't), backwards-in-time communication really REALLY F@#)(*s up causality. Heck, Faster Than Light (FTL) communication at all F@#)(*s up causality.
The EPR paradox, in particular, can't actually communicate information, as you can't actually tell which of the two sides measured it first (or at all), preventing FTL communication and saving causality. (Of course, simultinaity doesn't make sense either, thanks to relativity, but causality does).
This guy sounds like a good snake-oil salesman.
Rather than trying to archive everything, there is an important alternative, a data destruction policy. You can't discover what doesn't exist.
You just need to have the policy long in place before someone even thinks of suing you.
Any price cut will not be telegraphed in advance, as as soon as a price cut is announced, sales will plumet until the cut takes effect.
Long LONG before you can build a space elevator you need tether materials which are several orders of magnitude stronger than what we can build today...
If you could even get 1/100th of the way there on materials, you would have a great company selling fibers for military and industrial applications.
Power over Ethernet is for ordinary Cat 5e Gigabit Ethernet, it just checks for a magic resistor value and, if so, runs voltage over it.
PoE is a matter of switch choice, not cable choice.
I'd do the networking all as Cat5-e with Gigabit Ethernet...
Its a lot of bandwidth, cheap, and a universal lingua-franca.
I'd also have 802.11whosiwutzit access points, and more specifically cubbies with power so you can upgrade the access points.
Also, don't just string cable, string CONDUIT so you can upgrade the networking should you ever need to.
It looks like a subnotebook, it walks like a subnotebook, it bounces around in the briefcase like a subnotebook...
So why not get a real notebook?
A blatent ripoff add design specifically designed to cause confusion in the markeplace and violate trademarks was the target of apple's laywers.
A compromised card which is revoked but not used by an attacker COSTS money! Big money.
Its only if a compromised card is USED by an attacker that there is a problem. But since cards get stolen as well, they have heavy misuse detection to catch this, and if they let a few slip through, they aren't the ones holding the bill anyway, as it usually ends up being chargebacked to the merchant who accepted the bogus card.
A question: Do they still have net metering available (meter runs backwards if you overproduce)?
If so, does the net metering also respect the differential pricing?
This might actually be really useful then, as if you overbuild by 2x, those extra 4kW would be generated during the peak-price time.