OS/2 was heavily written by Microsoft in the early years (the 1.x versions were all called "Microsoft OS/2"); if Windows/386 hadn't been invented (setting the stage for Windows 3.0) and prompted the IBM-Microsoft Divorce, Microsoft would have still been the dominant PC OS vendor in the 1990s, selling OS/2. Which, given various factors, probably would have had a 16-bit 2.0 version similar to OS/2 1.x, and a 3.0 version based on the Windows NT kernel with a more OS/2-like 32-bit API.
However, let's assume Microsoft is wiped out by some sort of financial scandal in, oh, 1984. It ceases to exist, IBM winds up with exclusive control of PC/MS-DOS, Windows never comes along, IBM tries developing OS/2 in-house for the 286 processor.
Well, the most likely result here is the Revenge of Digital Research. DR ships DOS Plus (a predecessor of DR-DOS able to run both CP/M-86 and DOS 2.11 programs) and GEM/1 (a GUI) in 1985; the clonemakers buy both from DR instead of just DOS from IBM.
The likely evolution of PC OSes probably then follows the historical late 1980s evolution of DR products -- you wind up with a multitasking GEM (similar to the historical GEM/XM) and DOS (probably something similar to Concurrent DOS) pretty much filling the Windows 3.x role as everybody's standard x86 PC desktop, and an evolved version as Windows 95-equivalent. (Past there gets murky; does DR do a Windows NT? Do they use 4.4 BSD Lite and create a Unix that runs DOS/GEM programs? Or does a competitor knock them off the perch?)
Nuclear winter resulting in freezing cold and total collapse? The scenario existed in the original 1980s papers, but those models have long since been shown to be erroneous. Modern science shows that if you detonate 5000 megatons of nukes, spread over a thousand cities, you get a nuclear winter roughly as bad as the volcanic winter of 1816.
Er, no. The first time someone created document editing software that collapsed upon itself, sucked all surrounding code into it, and slowed the computer to a halt, they called it EMACS.
Sun went demanding license fees for its patent portfolio from NetApp. NetApp has responded with a lawsuit asking for the court to say that it isn't violating Sun's patents, and counterattacking by claiming Sun is violating NetApp patents. It's a classic defensive use of a patent portfolio against a bully, and I hope they rip Sun a new hole that just gushes money.
There's a huge difference in meaning between a section of a single definition which states "(including any such property intended to be attached to or installed in any real property without regard to whether it is so attached or installed)" and separate definition that says "or (2) anything designed or sold for incorporation into a dwelling". The text has gone from one definition to two, and "any real property" was replaced with "a dwelling". And most importantly, the text hasn't used the same wording.
If the GPL3 intended the meaning which interpretation has given Magnusson-Moss, they could have used the Magnusson-Moss wording straight. They didn't. Why would the GPL3 use its own definitions in place of Magnusson-Moss if it was intended to mean the same thing as Magnusson-Moss?
This becomes an especially important question given the cases that have turned on wording debates. Lawyers screw up their attempts to "get previous case law to apply to their new contracts/licenses" all the time, with millions of dollars in consequences. The result is that a careful lawyer would not change the wording unless another meaning was intended, because he'd know even a change intended as meaningless could result in changed interpretation. The fact that it was changed in wording is accordingly itself evidence in an argument that a different meaning was intended.
The GPL3 sues its own definitions for purposes of construing the contract. To quote the GPL3:
A "User Product" is either (1) a "consumer product", which means any tangible personal property which is normally used for personal, family, or household purposes, or (2) anything designed or sold for incorporation into a dwelling So, if a product isn't 'normally used for personal, family, or household purposes', and it isn't 'designed or sold for incorporation into a dwelling', it isn't covered by the anti-TiVoization clause. Dells sold through its business channel are 'normally used' for business purposes. So, a TiVoized Dell laptop running a GPL3 OS sold through Dell's business channel is immune from the anti-TiVoization clause; Dell can accordingly refuse to hand over the keys that would allow a modified version of the OS to run on the machine.
(In fact, given the dominance of Windows on the home desktop, vs. the use of Unix workalikes on desktops and servers for other purposes, one could make an argument that any general-purpose machine running a Unix workalike [other than OS X] is not 'normally used for personal, family, or household purposes', merely because of the OS used. While that might not hold alone, it would strengthen the argument that business-channel-only sales qualify as a non-'User Product' under the terms of the GPL3.)
To quote the GPL, instead of summarizing it:
A "User Product" is either (1) a "consumer product", which means any tangible personal property which is normally used for personal, family, or household purposes, or (2) anything designed or sold for incorporation into a dwelling So, if a product isn't 'normally used for personal, family, or household purposes', and it isn't 'designed or sold for incorporation into a dwelling', it isn't covered by the anti-TiVoization clause. Dells sold through its business channel are 'normally used' for business purposes. So, a TiVoized Dell laptop running a GPL3 OS sold through Dell's business channel is immune from the anti-TiVoization clause; Dell can refuse to hand over the keys that would allow a modified version of the OS to run on the machine.
The GPL3 doesn't do anything to protect the general-purpose computer. Dell could tomorrow sell all the TiVoized machines it likes, with half of them running a GPL3 OS, without disclosing anything. How? Just don't sell any of the ones with the GPL3 OS as "home" machines; just label them "business". Boom. The GPL3's own consumer-product-only rule lets Dell off the hook.
No, the only thing the GPL3 "anti-TiVoization" clause does is keep GPL3 software out of set-top boxes, home-targeted networking gear, and other devices designed as limited-use because they're for nontechnical users and making them hard to modify keeps down support costs. It's a big middle finger to TiVo which does nothing for software freedom.
These hacks are much bigger news for those outside America. Expect to see an industry spring up to meet European (and Asian?) demand for freed iPhones. I mean, hey, what European wouldn't want to pay northward of $600 for an obsolete (2.5 G) phone?
Under U.S. and international law, the U.S. Federal Government is effectively the owner of all spectrum longer than the near infrared in the U.S. There's a certain amount of international law, codified in treaties, dealing with certain segments, and broadcasts across borders, and such. There is then the frequencies the Federal Government directly uses or has reserved for its own use (mostly military). The FCC is the agency in charge of everything not allocated under international law or for Federal use.
The U.S. Congress has directed, by law, that the FCC auction off use rights to portions of the spectrum controlled by the FCC, as a method of allocating those frequencies (effectively government-owned) to non-governmental users. This is analogous to the U.S. Forest Service selling logging rights on federal land; the government owns the resource, Congress has set up a structure for sale of the right to use the resource through a designated agency, and the designated agency carries the sales out.
Wal-Mart was smaller than either K-mart or Target in the 1980s. How do you think they pulled ahead of both? Magic?
No, what Wal-Mart did was gather massive amounts of information, and learned to apply it better than the bigger chains. They worked out what sells and what didn't, kept their fingers on the pulse of demand, and then made their choices of what to stock and how accordingly. That upped their margins; they sold through inventory faster and thus made more profit on the same fixed store costs. That drove the profits that let them launch a huge wave of expansion -- and successfully compete with Target and K-Mart in the markets those two already dominated. Wal-Mart won because it had better information about what sells than its discount retail competitors.
And that's a huge reason for manufacturers to do what Wal-Mart tells them to do. Wal-Mart knows what sells at retail better than anybody else.
Take the example of Diet Coke with Splenda. Wal-Mart had better information on the demand for Splenda-sweetened soft drinks than Coca-Cola could have, because it knew how non-Coke Splenda soft drinks were selling. So when it told Coca-Cola that Splenda would be a good idea, Coca-Cola followed the advice -- and it's selling quite well, through all retail channels. Including, for example, Target.
If Wal-Mart says "jump", anyone who wants to make money will do so whether or not they care about selling through Wal-Mart, because Wal-Mart knows what the hell it's talking about.
Wow, it's just so weird to write "IBM/Solaris" Well, I've been expecting some form of Solaris on IBM ever since OpenSolaris came out, whether or not Sun was willing to cut a deal with IBM. How much can a Sun x86 Solaris box do that an IBM/OpenSolaris x86 box couldn't? How many Solaris-running SPARCs could be replaced with IBM/OpenSolaris POWER machines (either p or i series), if the price were right? The POWER/PowerPC OpenSolaris porting projects were going to (eventually) make IBM able to compete in Solaris shops that wouldn't dream of going Linux or AIX. How much marketshare would Sun lose to IBM? And how much share would have to be lost before people decided Sun was doomed as a company, and switched to other platforms (especially with the easy migration path of OpenSolaris on IBM)?
However, IBM has demonstrated with Linux it would rather let another company get a cut than go through the effort of building up a full support team for a new OS entirely in-house. Sun would rather get a cut of IBM/Solaris business than not get anything from IBM/OpenSolaris. So it's win-win; IBM gets lower costs, Sun gets money and security against the biggest danger presented by its decision to release OpenSolaris.
Prior to that, the closest thing was when NexGen (just before AMD bought them) developed an MMX-like extension for the Nx686 (released by AMD as the K6) and cut a deal for Cyrix to use them, which is what provoked Intel into creating MMX with cross-licensing to AMD and Cyrix.
Hmm. Best argument would be that Seattle Computer's QDOS illegally ripped off CP/M, putting MS-DOS and all its successors in violation. Which would make the claimant against Microsoft . . . DRDOS Inc., a spin-off of DeviceLogics, a spin-off of Lineo, a spin-off of Caldera Systems, which bought the rights from Novell, which bought Digital Research. (Caldera Systems, of course, is now known as The SCO Group.)
The Windows-ripped-off-MacOS and NT-ripped-off-VMS avenues are long-settled, but I don't know if the Caldera-vs-Microsoft settlement covered any lingering copyright infringement claims.
If it gets low enough, someone could snag it if for no other reason than to liquidate its assets. All of those desks, computer monitors, and coffee machines have got to be worth something. True. As of SCO's April 30th financials, SCO's Property/Plant/Equipment - Net (after depreciation) was worth $454,000.
That's all true. SCO might have anything from a small positive net value (if the SVRX fraction is low and they avoid owing anything on counterclaims) to a much more negative net value. And the book value on Reuters lists "--" for intangibles, so this doesn't count the value of their copyrights (they have some code they wrote and own; that's why the sales might not have been 100% SVRX) trademarks, and possible patents.
As of April 30, 2007, SCO had $19.847 million in total assets (which includes no intangibles), and $12.654 million in total liabilities, according to the Reuters data
The ruling on the Microsoft/Sun royalties owed to Novell adds roughly $25 million to the liabilities, making SCO worth roughly -18 million dollars, book value.
Re:Standing to bring suit
on
SCO Loses
·
· Score: 1
It possibly could, except the court also ruled "SCO is obligated to recognize Novell's waiver of SCO's claims against IBM and Sequent". SCO accordingly cannot make any copyright claims in its suit against IBM.
All this is is an opportunity to buy extra space for GMail/Picassa/etc. beyond what you already get on their servers for free. It is not an online storage service like Xdrive, but an equivalent to buying Hotmail Plus.
Okay, Valve distributes a copy of the software in violation of the license. It thus runs into: "Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License." So Valve's rights under the license are terminated.
So, how does Valve get out of this? Look at: "Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions."
So, Valve downloads a new copy of DOSBox. Upon completion of the download, under the "each time" clause, it automatically receives a new license, from the original licensor, to distribute the software under the GPL.
Now, yes, Valve can still be sued for copyright infringement on the copies it distributed in violation. And you might find a judge who won't rule "harmless oversight quickly repaired; no injury, so no damages." But further distribution is perfectly fine, since they're doing it under a new license issued by the original licensor, even though it has the same terms as the old one they breached.
Yeah, if somebody hits the Big Red Switch, there's going to be a problem. But, if they don't, well, it's a mainframe.
The Linux on these machines is running under z/VM, in multiple virtual machines. When one of them has a software fault, you reboot that one's VM and keep going; the other 132 Linux-running VMs run without noticing anything happened. (It is possible for z/VM to fault, sure. But it's an OS with 40 years of refinement in the "100% uptime" mainframe culture, and its task is just managing the virtual machines.) When something goes wrong with the hardware, the fault tolerance and self-healing features keep things running, and you fix the faulty element with a hot-swap. A properly set-up datacenter is going to minimize external risks, with backup power and such. Proper choice of datacenter location will minimize natural disaster risk.
So, yeah, the big risk is human failure, and these IBM-built, IBM-owned datacenters are presumably going to have extensively trained IBM-employed mainframe personnel, which minimizes that risk.
Now, if some cable company cuts the fiber optic lines . . .
Somebody directing a large effort does not have the ability to fully investigate every subsidiary dispute fairly without the effort grinding to a halt. Shorthand criteria are a necessity, as are rules of thumb as to what sources to bother investigating. This will inevitably lead to a number of less-than-optimum technical decisions along the way, but the results will be markedly superior to those where the manager stops everything to thoroughly investigate every aspect of every decision.
Con Kolivas's reaction to "losing" was not to continue to maintain SD and try to get it in later, or to try to improve CFS, but to quit kernel-hacking entirely. Which means he is not of a temperament that can accept that large projects will have arbitrary decisions that go against him, which means he would be a bad choice for the maintainer of a major kernel system. His actions in retrospect justify Torvalds's judgment that he couldn't trust him as a maintainer. Kolivas proved Torvalds correct on the management question, even if Torvalds is wrong on the technical one.
Slashdot Mirror
OS/2 was heavily written by Microsoft in the early years (the 1.x versions were all called "Microsoft OS/2"); if Windows/386 hadn't been invented (setting the stage for Windows 3.0) and prompted the IBM-Microsoft Divorce, Microsoft would have still been the dominant PC OS vendor in the 1990s, selling OS/2. Which, given various factors, probably would have had a 16-bit 2.0 version similar to OS/2 1.x, and a 3.0 version based on the Windows NT kernel with a more OS/2-like 32-bit API.
However, let's assume Microsoft is wiped out by some sort of financial scandal in, oh, 1984. It ceases to exist, IBM winds up with exclusive control of PC/MS-DOS, Windows never comes along, IBM tries developing OS/2 in-house for the 286 processor.
Well, the most likely result here is the Revenge of Digital Research. DR ships DOS Plus (a predecessor of DR-DOS able to run both CP/M-86 and DOS 2.11 programs) and GEM/1 (a GUI) in 1985; the clonemakers buy both from DR instead of just DOS from IBM.
The likely evolution of PC OSes probably then follows the historical late 1980s evolution of DR products -- you wind up with a multitasking GEM (similar to the historical GEM/XM) and DOS (probably something similar to Concurrent DOS) pretty much filling the Windows 3.x role as everybody's standard x86 PC desktop, and an evolved version as Windows 95-equivalent. (Past there gets murky; does DR do a Windows NT? Do they use 4.4 BSD Lite and create a Unix that runs DOS/GEM programs? Or does a competitor knock them off the perch?)
Nuclear winter resulting in freezing cold and total collapse? The scenario existed in the original 1980s papers, but those models have long since been shown to be erroneous. Modern science shows that if you detonate 5000 megatons of nukes, spread over a thousand cities, you get a nuclear winter roughly as bad as the volcanic winter of 1816.
New around here, aren't you?
Er, no. The first time someone created document editing software that collapsed upon itself, sucked all surrounding code into it, and slowed the computer to a halt, they called it EMACS.
Sun went demanding license fees for its patent portfolio from NetApp. NetApp has responded with a lawsuit asking for the court to say that it isn't violating Sun's patents, and counterattacking by claiming Sun is violating NetApp patents. It's a classic defensive use of a patent portfolio against a bully, and I hope they rip Sun a new hole that just gushes money.
Identical? Are we reading the same words?
There's a huge difference in meaning between a section of a single definition which states "(including any such property intended to be attached to or installed in any real property without regard to whether it is so attached or installed)" and separate definition that says "or (2) anything designed or sold for incorporation into a dwelling". The text has gone from one definition to two, and "any real property" was replaced with "a dwelling". And most importantly, the text hasn't used the same wording.
If the GPL3 intended the meaning which interpretation has given Magnusson-Moss, they could have used the Magnusson-Moss wording straight. They didn't. Why would the GPL3 use its own definitions in place of Magnusson-Moss if it was intended to mean the same thing as Magnusson-Moss?
This becomes an especially important question given the cases that have turned on wording debates. Lawyers screw up their attempts to "get previous case law to apply to their new contracts/licenses" all the time, with millions of dollars in consequences. The result is that a careful lawyer would not change the wording unless another meaning was intended, because he'd know even a change intended as meaningless could result in changed interpretation. The fact that it was changed in wording is accordingly itself evidence in an argument that a different meaning was intended.
(In fact, given the dominance of Windows on the home desktop, vs. the use of Unix workalikes on desktops and servers for other purposes, one could make an argument that any general-purpose machine running a Unix workalike [other than OS X] is not 'normally used for personal, family, or household purposes', merely because of the OS used. While that might not hold alone, it would strengthen the argument that business-channel-only sales qualify as a non-'User Product' under the terms of the GPL3.)
The GPL3 doesn't do anything to protect the general-purpose computer. Dell could tomorrow sell all the TiVoized machines it likes, with half of them running a GPL3 OS, without disclosing anything. How? Just don't sell any of the ones with the GPL3 OS as "home" machines; just label them "business". Boom. The GPL3's own consumer-product-only rule lets Dell off the hook.
No, the only thing the GPL3 "anti-TiVoization" clause does is keep GPL3 software out of set-top boxes, home-targeted networking gear, and other devices designed as limited-use because they're for nontechnical users and making them hard to modify keeps down support costs. It's a big middle finger to TiVo which does nothing for software freedom.
Under U.S. and international law, the U.S. Federal Government is effectively the owner of all spectrum longer than the near infrared in the U.S. There's a certain amount of international law, codified in treaties, dealing with certain segments, and broadcasts across borders, and such. There is then the frequencies the Federal Government directly uses or has reserved for its own use (mostly military). The FCC is the agency in charge of everything not allocated under international law or for Federal use.
The U.S. Congress has directed, by law, that the FCC auction off use rights to portions of the spectrum controlled by the FCC, as a method of allocating those frequencies (effectively government-owned) to non-governmental users. This is analogous to the U.S. Forest Service selling logging rights on federal land; the government owns the resource, Congress has set up a structure for sale of the right to use the resource through a designated agency, and the designated agency carries the sales out.
Wal-Mart was smaller than either K-mart or Target in the 1980s. How do you think they pulled ahead of both? Magic?
No, what Wal-Mart did was gather massive amounts of information, and learned to apply it better than the bigger chains. They worked out what sells and what didn't, kept their fingers on the pulse of demand, and then made their choices of what to stock and how accordingly. That upped their margins; they sold through inventory faster and thus made more profit on the same fixed store costs. That drove the profits that let them launch a huge wave of expansion -- and successfully compete with Target and K-Mart in the markets those two already dominated. Wal-Mart won because it had better information about what sells than its discount retail competitors.
And that's a huge reason for manufacturers to do what Wal-Mart tells them to do. Wal-Mart knows what sells at retail better than anybody else.
Take the example of Diet Coke with Splenda. Wal-Mart had better information on the demand for Splenda-sweetened soft drinks than Coca-Cola could have, because it knew how non-Coke Splenda soft drinks were selling. So when it told Coca-Cola that Splenda would be a good idea, Coca-Cola followed the advice -- and it's selling quite well, through all retail channels. Including, for example, Target.
If Wal-Mart says "jump", anyone who wants to make money will do so whether or not they care about selling through Wal-Mart, because Wal-Mart knows what the hell it's talking about.
However, IBM has demonstrated with Linux it would rather let another company get a cut than go through the effort of building up a full support team for a new OS entirely in-house. Sun would rather get a cut of IBM/Solaris business than not get anything from IBM/OpenSolaris. So it's win-win; IBM gets lower costs, Sun gets money and security against the biggest danger presented by its decision to release OpenSolaris.
x86-64 (AMD64) is the classic case.
Prior to that, the closest thing was when NexGen (just before AMD bought them) developed an MMX-like extension for the Nx686 (released by AMD as the K6) and cut a deal for Cyrix to use them, which is what provoked Intel into creating MMX with cross-licensing to AMD and Cyrix.
Hmm. Best argument would be that Seattle Computer's QDOS illegally ripped off CP/M, putting MS-DOS and all its successors in violation. Which would make the claimant against Microsoft . . . DRDOS Inc., a spin-off of DeviceLogics, a spin-off of Lineo, a spin-off of Caldera Systems, which bought the rights from Novell, which bought Digital Research. (Caldera Systems, of course, is now known as The SCO Group.)
The Windows-ripped-off-MacOS and NT-ripped-off-VMS avenues are long-settled, but I don't know if the Caldera-vs-Microsoft settlement covered any lingering copyright infringement claims.
That's all true. SCO might have anything from a small positive net value (if the SVRX fraction is low and they avoid owing anything on counterclaims) to a much more negative net value. And the book value on Reuters lists "--" for intangibles, so this doesn't count the value of their copyrights (they have some code they wrote and own; that's why the sales might not have been 100% SVRX) trademarks, and possible patents.
As of April 30, 2007, SCO had $19.847 million in total assets (which includes no intangibles), and $12.654 million in total liabilities, according to the Reuters data
The ruling on the Microsoft/Sun royalties owed to Novell adds roughly $25 million to the liabilities, making SCO worth roughly -18 million dollars, book value.
It possibly could, except the court also ruled "SCO is obligated to recognize Novell's waiver of SCO's claims against IBM and Sequent". SCO accordingly cannot make any copyright claims in its suit against IBM.
All this is is an opportunity to buy extra space for GMail/Picassa/etc. beyond what you already get on their servers for free. It is not an online storage service like Xdrive, but an equivalent to buying Hotmail Plus.
Okay, Valve distributes a copy of the software in violation of the license. It thus runs into: "Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License." So Valve's rights under the license are terminated.
So, how does Valve get out of this? Look at: "Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions."
So, Valve downloads a new copy of DOSBox. Upon completion of the download, under the "each time" clause, it automatically receives a new license, from the original licensor, to distribute the software under the GPL.
Now, yes, Valve can still be sued for copyright infringement on the copies it distributed in violation. And you might find a judge who won't rule "harmless oversight quickly repaired; no injury, so no damages." But further distribution is perfectly fine, since they're doing it under a new license issued by the original licensor, even though it has the same terms as the old one they breached.
Yeah, if somebody hits the Big Red Switch, there's going to be a problem. But, if they don't, well, it's a mainframe.
The Linux on these machines is running under z/VM, in multiple virtual machines. When one of them has a software fault, you reboot that one's VM and keep going; the other 132 Linux-running VMs run without noticing anything happened. (It is possible for z/VM to fault, sure. But it's an OS with 40 years of refinement in the "100% uptime" mainframe culture, and its task is just managing the virtual machines.) When something goes wrong with the hardware, the fault tolerance and self-healing features keep things running, and you fix the faulty element with a hot-swap. A properly set-up datacenter is going to minimize external risks, with backup power and such. Proper choice of datacenter location will minimize natural disaster risk.
So, yeah, the big risk is human failure, and these IBM-built, IBM-owned datacenters are presumably going to have extensively trained IBM-employed mainframe personnel, which minimizes that risk.
Now, if some cable company cuts the fiber optic lines . . .
Well, Grenada. It was a small war, yes, but people did get killed.
Fewer engineers and more people with degrees in Art History!
Somebody directing a large effort does not have the ability to fully investigate every subsidiary dispute fairly without the effort grinding to a halt. Shorthand criteria are a necessity, as are rules of thumb as to what sources to bother investigating. This will inevitably lead to a number of less-than-optimum technical decisions along the way, but the results will be markedly superior to those where the manager stops everything to thoroughly investigate every aspect of every decision.
Con Kolivas's reaction to "losing" was not to continue to maintain SD and try to get it in later, or to try to improve CFS, but to quit kernel-hacking entirely. Which means he is not of a temperament that can accept that large projects will have arbitrary decisions that go against him, which means he would be a bad choice for the maintainer of a major kernel system. His actions in retrospect justify Torvalds's judgment that he couldn't trust him as a maintainer. Kolivas proved Torvalds correct on the management question, even if Torvalds is wrong on the technical one.