35 U.S.C. 271 Infringement of patent. (a)Except as otherwise provided in this title, whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States, or imports into the United States any patented invention during the term of the patent therefor, infringes the patent.
There is no requirement to sell or otherwise have a commercial interest.
I don't know about other countries, but in the US that is flat out wrong. Anyone without authority who makes, uses, offers to sell, or sells any patented invention infringes the patent (35 USC 271). There is no 'it must be commercial' clause.
Well, once again you are wrong. Here is a link to a Computer World article from 1976. To save you some time, the headline is 'IBM, Amdahl Cross-license patents". The gist of the story is about how Amdahl will pay IBM for patent licenses, until such time that the value of Amdahl patents the IBM is using match the value of IBM patents that Amdahl is using. This was starting with Amdahl's very first machine.
IBM unbundled hardware and software, and started licensing (and charging for) software separately, in 1969. Amdahl was founded in 1970, and released it's first machine in 1975.
Note that this was in the 1970's when IBM was (unlike now) a monopoly, was under the 1956 consent decree, and was in the midst of the DOJ's (unsuccessful) anti-trust action. Amdahl STILL didn't get a 'free ride', like you seem to think you are entitled to.
SMB was invented as a way to interoperate between computers. And who invented it? Oh yeah, IBM. Microsoft took IBM's work and tried to make it Windows-only. It is hardly suprising that IBM opposed that. Now, are you going to try and claim that it was actually Hercules (or some other OSS project) that invented the z Architecture, and IBM then hijacked that?
Ah, so we're down to playing word games. Funny how you can so narrowly quote the pledge down to the word assert, but simply gloss over the fact that the whole pledge was only about a small subset of the tens of thousands of patents that IBM has. Yes, I know one or two of the patents listed on the pledge were on the list of 150 or so sent by Mr Anzani. Somehow I doubt that if they were not there it would make the slightest bit of difference in your statements.
By the way, any comparison to Google in their approach to patents is meaningless anyway. Google is primarily in the advertising business. IBM is mostly in the intellectual property business.
You and Maynard are the worst voices OSS could have. You give more ammo to the anti-OSS people than anyone else could. You actually do think is perfectly OK for someone to simply rip-off an investment representing of billions of dollars by investing nothing yourself, and then claim that is 'competition'. If someone painted OSS as that you would be jumping all over them, but that is exactly what you are doing. You have zero credibility left.
Failure to maintain and upgrade systems is not lock-in. It is just bad management. Similarly, choosing systems which have the performance and reliability characteristics you require for the money you want to spend is not lock-in. These two statements cover both your supposed airline 'lock-in' and the New Jersey 'lock-in'.Exactly how would Hercules (or any other OSS project) solve these problems?
And IBM has not violated their pledge in any way. They have not sued anyone. They have not sent a cease and desist letter to anyone. They have not even mentioned the existence of the patents to any member of the Hercules OSS project.
What would IBM possibly have to gain from suing anyone involved in the Hercules project? They have already taken all the action they need - they refused to license z/OS to run on it.
Neither IBM, nor Google, nor any other company who has made a similar statement has said they would actively SUPPORT a project that they felt infringed on their IP. All they have said is that they would take a passive approach and not sue. And that is exactly what they have done.
In the 10+ years of it's existence, has IBM ever sent any member of the Hercules project a cease and desist letter? Have they ever actually (not in someone's imagination) threatened a lawsuit? No. But you and a few others are trying to paint IBM's refusal to actively support a direct competitor as some sort of 'attack' on open source, and it is just plain dishonest.
They don't measure 'actual performance'. They measure time when z/OS was dispatching work. Using that and the known, stable performance characteristics of a particular model (and LPAR config) they calculate the charges. There are no known, stable performance characteristics of an emulated system. Even with workload charging there are still seven tiers of software pricing, depending on the performance of the machine.
If the license says 'run on IBM hardware', and you want it to run on non-IBM hardware, how is that not a change?
Also, what do you propose to do about the fact that z/OS (and all the IBM and ISV software that runs on it) is licensed and priced by the performance of the machine it is running on (which is impossible to predict on a FOSS emulator)? You not only want them to change the license, you want the whole industry to change their pricing models.
They didn't send anything to him AT ALL. They sent solicited private correspondence to a company he is supposedly not involved with, and that company made it public to start a smear campaign.
That first letter only talks about licensing z/OS to TurboHercules customers, which has absolutely nothing to do with patents. They were asking for permission to let non-IBM customers run z/OS. They were NOT asking IBM to grant licenses to emulate z/Architecture.
In the second letter to IBM, in the paragraph starting at the bottom of the first page, they make the quote I posted above
"In the unlikely event that IBM believes that the Hercules open source project infringes any IBM intellectual property, we kindly ask you too add any such property to the portfolio of patents that IBM has already pledged for the free use of the open source community".
So far you have denied both that the statement says they find it 'unlikely' that IBM would think it has IP in that area, and that they ask to be able to use said IP for free. We must be speaking different languages.
There is never a mention of any offer to license said IP on FRAND or any other terms.
And where, exactly, is the 'threat' to an OSS project that you and Maynard keep going on about? All any reasonable person sees in that exchange of letters is IBM declining to SUPPORT a direct competitor (not an OSS project), and, when pressed, give a detailed account of WHY they won't support them.
OK, so where did they ask IBM for FRAND licenses and get rejected? AFAIK, there has not been even a hint of a suggestion anywhere that they did that. Instead, they make statements like they think it is 'unlikely' that IBM, a company which proudly touts it's investments and patents, would have any IP relating to it's flagship hardware and architecture.
Are you kidding me? TH claimed they didn't even know there were patents until AFTER they filed the complaint. It seems far more likely that they filed the complaint, found out about the patents, and said 'we're screwed'. At that point they decided the best course of action was to get a few easily taken-in people to start a smear campaign against IBM.
I think you are entirely correct that a free lunch won't be achievable, so what FOSS license supports third-party patents?
I honestly have no idea what you are talking about anymore. A 2097-A01 costs approx $97000. An IFL costs $125000. I fail to see how an IFL is one tenth of the price, or what that has to do with anything anyway. Furthermore, how does artificially hobbling protect IBM's lock-in? Are you actually trying to say that these boxes (which IBM makes a profit on) are TOO competitive?
In another post you said that 'certainly no-one expects IBM to give it's IP away for free', and here you say they must 'retract their patent threats'. Well, which is it? Do you expect IBM to give away it's billions of dollars in investments or not?
Wait a sec, I thought your argument was that they don't offer a small mainframe, so you weren't taking their sales. Now it would appear as what you meant was they don't offer a CHEAP mainframe, and you do want to take their sales. Which is it?
"No one forgets that, and no one wants IBM to forget it. And most importantly, no one demands anything for free" Oh really? Is there an alternate explanation for this statement in the second letter to IBM?
"In the unlikely event that IBM believes that the Hercules open source project infringes any IBM intellectual property, we kindly ask you too add any such property to the portfolio of patents that IBM has already pledged for the free use of the open source community".
That statement seems to say to me quite clearly that they expect IBM to give up it's IP for FREE.
There are really only four possibilities for IBM:
1) Open z/OS licensing and ignore the patents - completely stupid business decision, absolutely nothing to gain for IBM, everything to lose 2) Open z/OS licensing and enforce the patents - sound business decision, no antitrust issues, doesn't help TurboHercules in the slightest, troubling for developers and users of Hercules 3) Keep z/OS licensing as is, and ignore the patents - sound business decision, possible anti-trust complaint, no harm to Hercules developers or users, TurboHercules out of luck 4) Keep z/OS licensing as is, and enforce the patents - possible anti-trust issues, bad PR, no real point
If there is another option, I would love to hear it.
Is your gigabyte of RAM at home in a system with a MTBF >30 years? Does your GB of RAM at home correct errors? Does it even DETECT errors? If it gets too many errors, does it silently fence off a section? If it fences off too many sections, does it automatically call the manufacturer to send out someone with a replacement? Once the replacement gets there, does the system and all applications continue running while the RAM is replaced? Is the cost of developing your non-correcting, non-detecting, non-hotswappable RAM spread out over hundreds of millions of customers, or only a few thousand? Gee, I can't imagine why mainframe RAM is more expensive, it must just be a rip-off.
The main problem is that you and TurboHercules are being entirely unreasonable. You do not only want IBM to license z/OS on other platforms, you want IBM to forget that they invested billions of dollars in the platform, and give the results of all that investment away for free.
Would you be happier if IBM opened the licensing for z/OS (ending the anti-trust claims), then turned around and sued for patent infringement?
Well, the other vendors that they licensed z/OS to also did this little thing called giving something in return, which Hercules/TurboHercules would just rather ignore. Exactly WHAT do you have to offer IBM? Cross-licenses for patents? License fees? Anything? No, the only thing you are 'offering' IBM is the opportunity to lose business. I can't imaging why they are not interested.
This is simply false. The smallest current generation 'mainframe' IBM sells is the 2098-A01, which runs at a paltry 28 MIPS, and can run any z/OS workload. The largest is the 2097-764, which runs at 30000 MIPS. In between those two extremes are several hundred more models to choose from. Almost any emulation is going to run better than 28 MIPS.
They are not patents that affect 'emulation', they are patents on elements of the z architecture. They do not come into play on the older OSs, because they don't run z architecture, and any patents on things used by those old OSs are long since expired.
As for the supposed anti-trust issues: what market is it that IBM supposedly dominates? Is it 'computing'? Is it 'servers'? Is it large-capacity servers? None of those seem to apply. The 'market', as you are defining it, is IBM z/Architecture machines. That is a pretty narrow definition of a market, isn't it? If that is your definition of a market, then every single company who has branded their products is a monopoly with no competition, and is subject to anti-trust actions.
You claim that IBM doesn't want to make z/OS available on non-IBM hardware. Do you have any evidence of that? I think all we know is that IBM declined to license z/OS on platforms which they believe infringe their patents. Was any effort made to resolve that issue? Did TurboHercules offer to license the patents, like any legitimate vendor would? No, instead they just demanded that IBM (after investing billions of dollars in them) give them away for free to someone who has invested exactly $0 and has nothing to offer IBM, except the opportunity to suck business away from IBM.
As for the NEON case - that is nothing more than a company that wants to sell a product allowing IBM's (and their ISV's) customers to violate their licensing agreements. If NEON was selling a product that allowed manufacturers to obscure their use of OSS code in violation of the licenses, what would your position be?
Can you provide us with a list of who these companies are that are making RECORD PROFITS every quarter? And what exactly is a RECORD PROFIT? Is it measured in dollars, or margin? Are they getting these profit increases from increased revenue, or decreased expense? What is the clip level for 'too much' profit? I seem to remember that the actual RECORD PROFIT was set by Exxon a few years ago, and it was something like $10B/qtr. What game company surpasses that?
Slashdot Mirror
Do you have a reference for that?
35 U.S.C. 271 Infringement of patent.
(a)Except as otherwise provided in this title, whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States, or imports into the United States any patented invention during the term of the patent therefor, infringes the patent.
There is no requirement to sell or otherwise have a commercial interest.
I don't know about other countries, but in the US that is flat out wrong. Anyone without authority who makes, uses, offers to sell, or sells any patented invention infringes the patent (35 USC 271). There is no 'it must be commercial' clause.
Well, once again you are wrong. Here is a link to a Computer World article from 1976. To save you some time, the headline is 'IBM, Amdahl Cross-license patents". The gist of the story is about how Amdahl will pay IBM for patent licenses, until such time that the value of Amdahl patents the IBM is using match the value of IBM patents that Amdahl is using. This was starting with Amdahl's very first machine.
IBM unbundled hardware and software, and started licensing (and charging for) software separately, in 1969. Amdahl was founded in 1970, and released it's first machine in 1975.
Note that this was in the 1970's when IBM was (unlike now) a monopoly, was under the 1956 consent decree, and was in the midst of the DOJ's (unsuccessful) anti-trust action. Amdahl STILL didn't get a 'free ride', like you seem to think you are entitled to.
SMB was invented as a way to interoperate between computers. And who invented it? Oh yeah, IBM. Microsoft took IBM's work and tried to make it Windows-only. It is hardly suprising that IBM opposed that. Now, are you going to try and claim that it was actually Hercules (or some other OSS project) that invented the z Architecture, and IBM then hijacked that?
Ah, so we're down to playing word games. Funny how you can so narrowly quote the pledge down to the word assert, but simply gloss over the fact that the whole pledge was only about a small subset of the tens of thousands of patents that IBM has. Yes, I know one or two of the patents listed on the pledge were on the list of 150 or so sent by Mr Anzani. Somehow I doubt that if they were not there it would make the slightest bit of difference in your statements.
By the way, any comparison to Google in their approach to patents is meaningless anyway. Google is primarily in the advertising business. IBM is mostly in the intellectual property business.
You and Maynard are the worst voices OSS could have. You give more ammo to the anti-OSS people than anyone else could. You actually do think is perfectly OK for someone to simply rip-off an investment representing of billions of dollars by investing nothing yourself, and then claim that is 'competition'. If someone painted OSS as that you would be jumping all over them, but that is exactly what you are doing. You have zero credibility left.
Failure to maintain and upgrade systems is not lock-in. It is just bad management. Similarly, choosing systems which have the performance and reliability characteristics you require for the money you want to spend is not lock-in. These two statements cover both your supposed airline 'lock-in' and the New Jersey 'lock-in' .Exactly how would Hercules (or any other OSS project) solve these problems?
And IBM has not violated their pledge in any way. They have not sued anyone. They have not sent a cease and desist letter to anyone. They have not even mentioned the existence of the patents to any member of the Hercules OSS project.
What would IBM possibly have to gain from suing anyone involved in the Hercules project? They have already taken all the action they need - they refused to license z/OS to run on it.
Neither IBM, nor Google, nor any other company who has made a similar statement has said they would actively SUPPORT a project that they felt infringed on their IP. All they have said is that they would take a passive approach and not sue. And that is exactly what they have done.
In the 10+ years of it's existence, has IBM ever sent any member of the Hercules project a cease and desist letter? Have they ever actually (not in someone's imagination) threatened a lawsuit? No. But you and a few others are trying to paint IBM's refusal to actively support a direct competitor as some sort of 'attack' on open source, and it is just plain dishonest.
They don't measure 'actual performance'. They measure time when z/OS was dispatching work. Using that and the known, stable performance characteristics of a particular model (and LPAR config) they calculate the charges. There are no known, stable performance characteristics of an emulated system. Even with workload charging there are still seven tiers of software pricing, depending on the performance of the machine.
If the license says 'run on IBM hardware', and you want it to run on non-IBM hardware, how is that not a change?
Also, what do you propose to do about the fact that z/OS (and all the IBM and ISV software that runs on it) is licensed and priced by the performance of the machine it is running on (which is impossible to predict on a FOSS emulator)? You not only want them to change the license, you want the whole industry to change their pricing models.
They didn't send anything to him AT ALL. They sent solicited private correspondence to a company he is supposedly not involved with, and that company made it public to start a smear campaign.
Did you ask Google's lawyer if they would support a direct competitor, who happens to be using an OSS project that infringes Googles patents?
That first letter only talks about licensing z/OS to TurboHercules customers, which has absolutely nothing to do with patents. They were asking for permission to let non-IBM customers run z/OS. They were NOT asking IBM to grant licenses to emulate z/Architecture.
In the second letter to IBM, in the paragraph starting at the bottom of the first page, they make the quote I posted above
"In the unlikely event that IBM believes that the Hercules open source project infringes any IBM intellectual property, we kindly ask you too add any such property to the portfolio of patents that IBM has already pledged for the free use of the open source community".
So far you have denied both that the statement says they find it 'unlikely' that IBM would think it has IP in that area, and that they ask to be able to use said IP for free. We must be speaking different languages.
There is never a mention of any offer to license said IP on FRAND or any other terms.
And where, exactly, is the 'threat' to an OSS project that you and Maynard keep going on about? All any reasonable person sees in that exchange of letters is IBM declining to SUPPORT a direct competitor (not an OSS project), and, when pressed, give a detailed account of WHY they won't support them.
OK, so where did they ask IBM for FRAND licenses and get rejected? AFAIK, there has not been even a hint of a suggestion anywhere that they did that. Instead, they make statements like they think it is 'unlikely' that IBM, a company which proudly touts it's investments and patents, would have any IP relating to it's flagship hardware and architecture.
Are you kidding me? TH claimed they didn't even know there were patents until AFTER they filed the complaint. It seems far more likely that they filed the complaint, found out about the patents, and said 'we're screwed'. At that point they decided the best course of action was to get a few easily taken-in people to start a smear campaign against IBM.
I think you are entirely correct that a free lunch won't be achievable, so what FOSS license supports third-party patents?
I honestly have no idea what you are talking about anymore. A 2097-A01 costs approx $97000. An IFL costs $125000. I fail to see how an IFL is one tenth of the price, or what that has to do with anything anyway. Furthermore, how does artificially hobbling protect IBM's lock-in? Are you actually trying to say that these boxes (which IBM makes a profit on) are TOO competitive?
In another post you said that 'certainly no-one expects IBM to give it's IP away for free', and here you say they must 'retract their patent threats'. Well, which is it? Do you expect IBM to give away it's billions of dollars in investments or not?
Wait a sec, I thought your argument was that they don't offer a small mainframe, so you weren't taking their sales. Now it would appear as what you meant was they don't offer a CHEAP mainframe, and you do want to take their sales. Which is it?
"No one forgets that, and no one wants IBM to forget it. And most importantly, no one demands anything for free" Oh really? Is there an alternate explanation for this statement in the second letter to IBM?
"In the unlikely event that IBM believes that the Hercules open source project infringes any IBM intellectual property, we kindly ask you too add any such property to the portfolio of patents that IBM has already pledged for the free use of the open source community".
That statement seems to say to me quite clearly that they expect IBM to give up it's IP for FREE.
There are really only four possibilities for IBM:
1) Open z/OS licensing and ignore the patents - completely stupid business decision, absolutely nothing to gain for IBM, everything to lose
2) Open z/OS licensing and enforce the patents - sound business decision, no antitrust issues, doesn't help TurboHercules in the slightest, troubling for developers and users of Hercules
3) Keep z/OS licensing as is, and ignore the patents - sound business decision, possible anti-trust complaint, no harm to Hercules developers or users, TurboHercules out of luck
4) Keep z/OS licensing as is, and enforce the patents - possible anti-trust issues, bad PR, no real point
If there is another option, I would love to hear it.
Is your gigabyte of RAM at home in a system with a MTBF >30 years? Does your GB of RAM at home correct errors? Does it even DETECT errors? If it gets too many errors, does it silently fence off a section? If it fences off too many sections, does it automatically call the manufacturer to send out someone with a replacement? Once the replacement gets there, does the system and all applications continue running while the RAM is replaced? Is the cost of developing your non-correcting, non-detecting, non-hotswappable RAM spread out over hundreds of millions of customers, or only a few thousand? Gee, I can't imagine why mainframe RAM is more expensive, it must just be a rip-off.
The main problem is that you and TurboHercules are being entirely unreasonable. You do not only want IBM to license z/OS on other platforms, you want IBM to forget that they invested billions of dollars in the platform, and give the results of all that investment away for free.
Would you be happier if IBM opened the licensing for z/OS (ending the anti-trust claims), then turned around and sued for patent infringement?
Well, the other vendors that they licensed z/OS to also did this little thing called giving something in return, which Hercules/TurboHercules would just rather ignore. Exactly WHAT do you have to offer IBM? Cross-licenses for patents? License fees? Anything? No, the only thing you are 'offering' IBM is the opportunity to lose business. I can't imaging why they are not interested.
This is simply false. The smallest current generation 'mainframe' IBM sells is the 2098-A01, which runs at a paltry 28 MIPS, and can run any z/OS workload. The largest is the 2097-764, which runs at 30000 MIPS. In between those two extremes are several hundred more models to choose from. Almost any emulation is going to run better than 28 MIPS.
They are not patents that affect 'emulation', they are patents on elements of the z architecture. They do not come into play on the older OSs, because they don't run z architecture, and any patents on things used by those old OSs are long since expired.
As for the supposed anti-trust issues: what market is it that IBM supposedly dominates? Is it 'computing'? Is it 'servers'? Is it large-capacity servers? None of those seem to apply. The 'market', as you are defining it, is IBM z/Architecture machines. That is a pretty narrow definition of a market, isn't it? If that is your definition of a market, then every single company who has branded their products is a monopoly with no competition, and is subject to anti-trust actions.
You claim that IBM doesn't want to make z/OS available on non-IBM hardware. Do you have any evidence of that? I think all we know is that IBM declined to license z/OS on platforms which they believe infringe their patents. Was any effort made to resolve that issue? Did TurboHercules offer to license the patents, like any legitimate vendor would? No, instead they just demanded that IBM (after investing billions of dollars in them) give them away for free to someone who has invested exactly $0 and has nothing to offer IBM, except the opportunity to suck business away from IBM.
As for the NEON case - that is nothing more than a company that wants to sell a product allowing IBM's (and their ISV's) customers to violate their licensing agreements. If NEON was selling a product that allowed manufacturers to obscure their use of OSS code in violation of the licenses, what would your position be?
Can you provide us with a list of who these companies are that are making RECORD PROFITS every quarter? And what exactly is a RECORD PROFIT? Is it measured in dollars, or margin? Are they getting these profit increases from increased revenue, or decreased expense? What is the clip level for 'too much' profit? I seem to remember that the actual RECORD PROFIT was set by Exxon a few years ago, and it was something like $10B/qtr. What game company surpasses that?
Oak Ridge is in the TVA, which is hydro, not coal. The lab is there precisely because of the available power.