If you designed a product and one of the components of that product was an iPhone, literally integrated into the product itself as part of a larger whole, would you pay licensing fees to Qualcomm?
I ask this because that's what Qualcomm would expect you to do, although the manufacturers of the chips in that iPhone have already paid licensing fees and, if Qualcomm gets their way, so will Apple. So they'll have collected licensing fees on, say, a $10 chip and an $800 phone, and will be trying to collect fees from you on what may be a $2000 device; so they're, effectively, trying to collect licensing fees based on a $2810 sale price for an item that eventually sells for $2000. Sorry, but no.
I was stating my callback policy for the benefit of those who might receive calls from scammers, not because my banks ask for personal info on cold calls; because they don't.
It doesn't matter that an actual call from your bank will "tell you that you need to call them, using the number printed on the back of your Credit Card" if it's not your bank calling and you're a member of the majority of the population who have too much going on in their lives to take the time to inquire as to their bank's fraud prevention procedures. Without a personal policy in place regarding handling of personal and account details, anyone with a bank account at any bank is vulnerable to cold-call scams.
You'd probably know if your bank had such a system, as the damn thing has a hair trigger. One of my cards gets declined if I try to used it more than 20 miles or so from home and I have to either use a different card or step out of line, wait for the fraud call, bitch at them loudly enough that everyone staring at me like a deadbeat can hear it, then try the transaction again. As a result, i use that card less than the others, as the other cards will let the transaction go through, then call.
Your bank makes you fill out and sign a form for that? Shit, I have cards with 4 different banks (5 if you count store cards through Synchrony, but I've never dealt with fraud on those cards) and all 4 of them call me if they think a transaction is fraudulent; I can press 1 to verify the transaction or 2 to speak with a representative about potential fraud. All 4 banks I deal with have this system, and it works well; I've encountered actual fraud on cards from each of those banks and it's always been a matter of simply telling them the transaction wasn't initiated by me. Transaction immediately reversed, new card arrives in 2 days, nothing to fill out or sign.
Of course, due diligence, when they call I refuse to provide any account details and I check my account online to ensure what they're saying matches what I'm seeing before I confirm or deny anything. If they ask for account details, I ask their extension and tell them I'll call back at the number on the back of my card. Never provide personal information or payment/account details on an incoming call, because caller ID can be faked.
In all honesty, I think the solution lies somewhere between your concept and mine. Injunction should be an option where licensing fails; without that, what incentive does a large corporation have to accept the little guy's licensing terms?
So, then, Apple has a contract with Qualcomm that stipulates they should be paying a licensing fee per-unit?
Again, Apple isn't so stupid as to take that to court where they know they'll lose.
We'll surely learn more when that time comes.
But, until then, let me leave you with this:
The "programmable streaming processor" described in 8,633,936 is a GPU, it runs the program loaded on it at runtime by whatever software is using the display. It isn't fatcory-programmed and, in fact, mentions "a non-transitory computer-readable storage medium"; if you know what "transitory" means, you also know that "non-transitory" in this context, means RAM. Firmware, by its nature, lives in transitory storage, and the described chip runs a program from non-transitory storage. While this sort of processor may contain firmware (in the form of microcode), there is no mention of it in this patent. This part of the complaint relates directly to a physical chip.
The "envelope tracker" described in 8,698,558 "generates a second supply voltage based on an envelope signal and the boosted supply voltage", it relies in its current inputs to determine its current outputs and does not use firmware. Do you know what an envelope amplifier is? Yes, there are envelope amplifiers which are programmed by firmware, but this one is not. This part of the complaint relates directly to a physical chip.
The "level shifter" described in 8,487,658 is a "layout design". You really don't even have to read the patent to know it doesn't involve firmware, as I gleaned all of that from the title, but here goes... Just from the abstract, we get mentions of the specific construction of the silicon ("voltage level shifters (VLS) design in bulk CMOS technology") and a description of function of a number of N substrates of a multi-transistor chip. There's no such thing as CMOS firmware (firmware can reside on a CMOS chip, but that's a different concept from the one described within this patent) and firmware does not have N substrates. I mean, come on, did you even read these patents? Let's look at some of the claims... "A multi-voltage circuit to shift each of two bits from a first voltage level logic to a second voltage level logic, comprising:" (scans for mention of any means of storing or loading firmware -- finds none), "A four-bit multi-voltage circuit to shift each of four bits from a first voltage level logic to a second voltage level logic, comprising:" (a 4-bit register and some N substrates, but no means of including firmware), "A method for reducing die area" (which certainly isn't implemented in firmware -- and repeated for the remainder of the claims in the patent, though they do switch between "method" and "apparatus"). No firmware in this chip, either. This part of the complaint relates directly to a physical chip.
Then, there's 8,838,949, which describes, of all things, "A multi-processor system comprising [a secondary processor], [a scatter loader controller], [a primary processor coupled with a memory], and [an interface communicatively coupling the primary processor and the secondary processor]". I'll admit, it does mention the scatter loader controller being configured "to load the image header; and to scatter load each received data segment based at least in part on the loaded image header, directly from the hardware buffer to the system memory", which sounds like, maybe, some of that firmware you're going on about. Let's look deeper, shall we? Well, no mention of the mechanism for configuration over the 8 subclaims to claim 1, which brings us to claim 10 which, well, let's just say doesn't mention how the device is configured, either. Neither do its sub-claims, so, on to claim 16 and it's single sub-claim, which also make no mention of how the unit should be configured. Ooh! Claims 18 and 20 mention non-volatile memory for storing programs for the primary and secondary processors! It looks like we finally found the firmware you're talking abo
most people just license up for the firmware and options they want. Except Apple
Again: Where is Apple getting Qualcomm's firmware if not from Qualcomm? Seems if they're getting it from Qualcomm they must've paid for it and, if they're getting it from elsewhere, that source should be named in the complaint, as well.
But let's look at those patents, shall we? (and thanks, I hadn't seen this previously)
Patent 8,633,936, entitled "Programmable streaming processor with mixed precision instruction execution", is for a silicon design. That is, it's for a chip, not firmware.
Patent 8,698,558, entitled "Low-voltage power-efficient envelope tracker", is for a silicon design. Again, that's a chip, not firmware.
Patent 8,487,658, entitled "Compact and robust level shifter layout design", is for... need I even say it? It's a silicon design, another chip, still not firmware.
Patent 8,838,949, entitled "Direct scatter loading of executable software image from a primary processor to one or more secondary processor in a multi-processor system", describes a method for loading a binary image from one CPU to another, directly in....... of all things, would you believe it's silicon? Still no firmware.
Patent 9,535,490, entitled "Power saving techniques in computing devices". describes a system for enabling and disabling system buses (again, in silicon) based on presence of data to be passed over said buses, as a means of power conservation. Still no firmware.
Patent 9,608,675 is the last patent in the complaint. This must be where the firmware is, right? Let us look, let us look.
Well, would you look at that, it's entitled "Power tracker for multiple transmit signals sent simultaneously" and describes an arrangement of discreet circuits which may also be implemented directly in what is rapidly becoming one of my favorite materials: silicon.
So, I count 6 total patents in the complaint. Of those, 6 describe silicon and circuitry and 0 describe firmware.
Only one of us is talking out our ass...
I'm not sure what hole you think you're talking out of but, having read the patents referenced in the complaint... well, you're right, only one of us... and that would be you, my friend.
When did you last try and were you fully updated (Windows and Ubuntu)? On the initial release a lot of stuff was broken, but MS was on top of fixing as much of what was reported as they could in the next release, then fixed more in the next release. The few projects I've built in the current release worked fine.
You do realize that doesn't apply to all chips, right? And, again, where are they getting the firmware?
My understanding is that Qualcomm is bitching about patents relating to silicon, not software. Of course, none of that is actually public, which is why we're both talking out our asses right now and will have to wait to hear Qualcomm's actual complaints in court.
but he doesn't get to submarine other's work 15 years down the line, like happens today with trolls.
This is why I stipulated monthly progress updates (which must show actual progress) until ac working model has been created. If progress stops for a period (we'll say 3 months), the patent is lost; if someone manages to keep up real and provable progress for 15 years without actually producing anything, well, I'd be amazed.
I also believe the patent holder should be able to file an injunction and prevent the manufacture, import, and sale of their invention while they're building their working model. I'm all for requiring the patent holder to product and present a functioning unit in order to collect damages, though. I think that's a fair enough compromise; it would certainly stop patent trolls dead in their tracks as they would no longer be able to profit from holding up innovation. That is the goal, right?
And where are they getting the firmware, if not from Qualcomm? And if they're not getting firmware from Qualcomm, they're using 3rd-party firmware and that 3rd party would be responsible for licensing unless their contract with Apple says otherwise, in which case it would fall to Apple and Apple wouldn't be so stupid as to take that fight to court where they know they'd lose.
Clearly, if Apple is fighting this, the license fees have been paid somewheresomeone. Qualcomm is too big for Apple to bankrupt with a drawn-out court battle; yes, technically they have the money to do so but, as a publicly traded company, their board of directors would have to answer for the massive dip in profits over however many quarters that took.
At any rate, we're both just talking out our asses right now. Since it looks like Apple intends to take this to court, though, we'll get to see soon enough which of us was right.
Yes, it does address trolls quite well. What it doesn't address is the sole inventor who can only work on his inventions in his spare time because he has bills to pay; how does anything you've proposed thus far protect him from some corporate entity learning what he's working on (maybe he shares a few details with a friend in a bar and someone overhears, for example) and they get the thing built in a week, while he's still months, or a year or more, from finishing?
What you propose would be great for corporations who actually produce the things they invent, but not so great for individuals who may also wish to produce the things they invent, but have other obligations and, so, can't do so as quickly.
I'd go almost that far, but you must consider the time it might take a single inventor on no budget to build his invention, versus the time it would take a corporate-backed team. I'd say you must actively be building it, and that periodic proof of progress (until completion) should be required in order to have an enforceable patent. This would protect the little guy who might take a year or longer working part-time on building his invention (after all, he has bills to pay) from larger competitors who may otherwise steal his idea by being first to build (after all, they have budgets for R&D, corporate espionage, and manufacturing, as well as staff on hand to work on it full-time in a team capacity).
If the suppliers stopped paying, Apple should pay licensing for all parts acquired for which the supplier did not. However, if Apple asked suppliers to stop paying and the suppliers did not stop, well, licensing is still being paid and Apple doesn't owe shit.
Well, if Apple is being expected to pay, that's only fair. In fact, Qualcomm should welcome that, as their licensing is based on the final sale price of the finished product; it would be in Qualcomm's best interest to charge their licensing fee as late in the chain as possible.
You do realize that you can now run an full Ubuntu environment within Windows, right? Last I checked, Windows will run Word and Excel natively, too. Cygwin is so 2015.
Yes, one built last month on the AM3+ platform. That's an AMD FX CPU, at least as old as the Mac it's being compared to. Let's see... 8 core... the newest one of those is hmm... the FX-9590, released in 2013.
That's four years old.
Funny, that's the same benchmark you set for beating the three year old Mac.
Because the manufacturers of the chips the use in their phones have already paid those several dollars per unit per patent and passed that cost along to Apple in their unit pricing. In short, Apple has already paid it once, when they bought the chips, and rightly does not feel they should have to pay it again.
This. Especially since the chip manufacturer would already have paid Qualcomm to license their technology, otherwise the chip manufacturer would be the one being sued or Apple would be willing to pay Qualcomm directly as part of their agreement with said chip manufacturer.
Apple continues to use Qualcomm's technology while refusing to pay for it twice
Someone makes the chips Apple uses in their phones. Those chips implement Qualcomm's patented technologies and whoever makes them, therefore, must be licensing those technologies, which means they're paying for them and passing that cost along to Apple who, by paying for the chips which incorporate Qualcomm's technologies, made by the company who already paid for Qualcomm's technologies, has already paid for the use of Qualcomm's technologies.
I realize that's hard for some people (namely Qualcomm's leadership and council) to follow, so let's look at a similar situation with a different type of product.
Coca-Cola uses high fructose corn syrup in their products. They buy this high fructose corn syrup from a supplier. They pay for this high fructose corn syrup when they buy it, then they incorporate it into their products, which they then sell to stores. The stores, then, sell the products to consumers. The stores do not owe the high fructose corn syrup manufacturer anything for the use and sale of their product, because Coca-Cola already paid for that.
Rephrased to fit the Qualcomm situation, with edits bolded to make them clear:
Chip manufacturers use Qualcomm technologies in their products. They license these Qualcomm technologies from a Qualcomm. They pay for these Qualcomm technologies when they license them, then they incorporate them into their products, which they then sell to Apple. Apple, then, sells the products to consumers. Apple does not owe Qualcomm anything for the use and sale of their technologies, because the chip manufacturers already paid for that.
Slashdot Mirror
All that would lead to is even shittier products costing even more and being even shinier to get people to still buy them.
So, then, let me ask you this:
If you designed a product and one of the components of that product was an iPhone, literally integrated into the product itself as part of a larger whole, would you pay licensing fees to Qualcomm?
I ask this because that's what Qualcomm would expect you to do, although the manufacturers of the chips in that iPhone have already paid licensing fees and, if Qualcomm gets their way, so will Apple. So they'll have collected licensing fees on, say, a $10 chip and an $800 phone, and will be trying to collect fees from you on what may be a $2000 device; so they're, effectively, trying to collect licensing fees based on a $2810 sale price for an item that eventually sells for $2000. Sorry, but no.
Qualcomm is double-dipping, plain and simple.
I was stating my callback policy for the benefit of those who might receive calls from scammers, not because my banks ask for personal info on cold calls; because they don't.
It doesn't matter that an actual call from your bank will "tell you that you need to call them, using the number printed on the back of your Credit Card" if it's not your bank calling and you're a member of the majority of the population who have too much going on in their lives to take the time to inquire as to their bank's fraud prevention procedures. Without a personal policy in place regarding handling of personal and account details, anyone with a bank account at any bank is vulnerable to cold-call scams.
You'd probably know if your bank had such a system, as the damn thing has a hair trigger. One of my cards gets declined if I try to used it more than 20 miles or so from home and I have to either use a different card or step out of line, wait for the fraud call, bitch at them loudly enough that everyone staring at me like a deadbeat can hear it, then try the transaction again. As a result, i use that card less than the others, as the other cards will let the transaction go through, then call.
Your bank makes you fill out and sign a form for that? Shit, I have cards with 4 different banks (5 if you count store cards through Synchrony, but I've never dealt with fraud on those cards) and all 4 of them call me if they think a transaction is fraudulent; I can press 1 to verify the transaction or 2 to speak with a representative about potential fraud. All 4 banks I deal with have this system, and it works well; I've encountered actual fraud on cards from each of those banks and it's always been a matter of simply telling them the transaction wasn't initiated by me. Transaction immediately reversed, new card arrives in 2 days, nothing to fill out or sign.
Of course, due diligence, when they call I refuse to provide any account details and I check my account online to ensure what they're saying matches what I'm seeing before I confirm or deny anything. If they ask for account details, I ask their extension and tell them I'll call back at the number on the back of my card. Never provide personal information or payment/account details on an incoming call, because caller ID can be faked.
By the time I pull my knife on you, I'm already in too close for you to draw your gun.
And yes, I carry both.
For that very reason.
Indeed, I did mean net, and I need to do a better job proofreading before I post.
In all honesty, I think the solution lies somewhere between your concept and mine. Injunction should be an option where licensing fails; without that, what incentive does a large corporation have to accept the little guy's licensing terms?
So, then, Apple has a contract with Qualcomm that stipulates they should be paying a licensing fee per-unit?
Again, Apple isn't so stupid as to take that to court where they know they'll lose.
We'll surely learn more when that time comes.
But, until then, let me leave you with this:
The "programmable streaming processor" described in 8,633,936 is a GPU, it runs the program loaded on it at runtime by whatever software is using the display. It isn't fatcory-programmed and, in fact, mentions "a non-transitory computer-readable storage medium"; if you know what "transitory" means, you also know that "non-transitory" in this context, means RAM. Firmware, by its nature, lives in transitory storage, and the described chip runs a program from non-transitory storage. While this sort of processor may contain firmware (in the form of microcode), there is no mention of it in this patent. This part of the complaint relates directly to a physical chip.
The "envelope tracker" described in 8,698,558 "generates a second supply voltage based on an envelope signal and the boosted supply voltage", it relies in its current inputs to determine its current outputs and does not use firmware. Do you know what an envelope amplifier is? Yes, there are envelope amplifiers which are programmed by firmware, but this one is not. This part of the complaint relates directly to a physical chip.
The "level shifter" described in 8,487,658 is a "layout design". You really don't even have to read the patent to know it doesn't involve firmware, as I gleaned all of that from the title, but here goes... Just from the abstract, we get mentions of the specific construction of the silicon ("voltage level shifters (VLS) design in bulk CMOS technology") and a description of function of a number of N substrates of a multi-transistor chip. There's no such thing as CMOS firmware (firmware can reside on a CMOS chip, but that's a different concept from the one described within this patent) and firmware does not have N substrates. I mean, come on, did you even read these patents? Let's look at some of the claims... "A multi-voltage circuit to shift each of two bits from a first voltage level logic to a second voltage level logic, comprising:" (scans for mention of any means of storing or loading firmware -- finds none), "A four-bit multi-voltage circuit to shift each of four bits from a first voltage level logic to a second voltage level logic, comprising:" (a 4-bit register and some N substrates, but no means of including firmware), "A method for reducing die area" (which certainly isn't implemented in firmware -- and repeated for the remainder of the claims in the patent, though they do switch between "method" and "apparatus"). No firmware in this chip, either. This part of the complaint relates directly to a physical chip.
Then, there's 8,838,949, which describes, of all things, "A multi-processor system comprising [a secondary processor], [a scatter loader controller], [a primary processor coupled with a memory], and [an interface communicatively coupling the primary processor and the secondary processor]". I'll admit, it does mention the scatter loader controller being configured "to load the image header; and to scatter load each received data segment based at least in part on the loaded image header, directly from the hardware buffer to the system memory", which sounds like, maybe, some of that firmware you're going on about. Let's look deeper, shall we? Well, no mention of the mechanism for configuration over the 8 subclaims to claim 1, which brings us to claim 10 which, well, let's just say doesn't mention how the device is configured, either. Neither do its sub-claims, so, on to claim 16 and it's single sub-claim, which also make no mention of how the unit should be configured. Ooh! Claims 18 and 20 mention non-volatile memory for storing programs for the primary and secondary processors! It looks like we finally found the firmware you're talking abo
most people just license up for the firmware and options they want. Except Apple
Again: Where is Apple getting Qualcomm's firmware if not from Qualcomm? Seems if they're getting it from Qualcomm they must've paid for it and, if they're getting it from elsewhere, that source should be named in the complaint, as well.
But let's look at those patents, shall we? (and thanks, I hadn't seen this previously)
Patent 8,633,936, entitled "Programmable streaming processor with mixed precision instruction execution", is for a silicon design. That is, it's for a chip, not firmware.
Patent 8,698,558, entitled "Low-voltage power-efficient envelope tracker", is for a silicon design. Again, that's a chip, not firmware.
Patent 8,487,658, entitled "Compact and robust level shifter layout design", is for... need I even say it? It's a silicon design, another chip, still not firmware.
Patent 8,838,949, entitled "Direct scatter loading of executable software image from a primary processor to one or more secondary processor in a multi-processor system", describes a method for loading a binary image from one CPU to another, directly in....... of all things, would you believe it's silicon? Still no firmware.
Patent 9,535,490, entitled "Power saving techniques in computing devices". describes a system for enabling and disabling system buses (again, in silicon) based on presence of data to be passed over said buses, as a means of power conservation. Still no firmware.
Patent 9,608,675 is the last patent in the complaint. This must be where the firmware is, right? Let us look, let us look.
Well, would you look at that, it's entitled "Power tracker for multiple transmit signals sent simultaneously" and describes an arrangement of discreet circuits which may also be implemented directly in what is rapidly becoming one of my favorite materials: silicon.
So, I count 6 total patents in the complaint. Of those, 6 describe silicon and circuitry and 0 describe firmware.
Only one of us is talking out our ass...
I'm not sure what hole you think you're talking out of but, having read the patents referenced in the complaint... well, you're right, only one of us... and that would be you, my friend.
When did you last try and were you fully updated (Windows and Ubuntu)? On the initial release a lot of stuff was broken, but MS was on top of fixing as much of what was reported as they could in the next release, then fixed more in the next release. The few projects I've built in the current release worked fine.
You do realize that doesn't apply to all chips, right? And, again, where are they getting the firmware?
My understanding is that Qualcomm is bitching about patents relating to silicon, not software. Of course, none of that is actually public, which is why we're both talking out our asses right now and will have to wait to hear Qualcomm's actual complaints in court.
but he doesn't get to submarine other's work 15 years down the line, like happens today with trolls.
This is why I stipulated monthly progress updates (which must show actual progress) until ac working model has been created. If progress stops for a period (we'll say 3 months), the patent is lost; if someone manages to keep up real and provable progress for 15 years without actually producing anything, well, I'd be amazed.
I also believe the patent holder should be able to file an injunction and prevent the manufacture, import, and sale of their invention while they're building their working model. I'm all for requiring the patent holder to product and present a functioning unit in order to collect damages, though. I think that's a fair enough compromise; it would certainly stop patent trolls dead in their tracks as they would no longer be able to profit from holding up innovation. That is the goal, right?
And where are they getting the firmware, if not from Qualcomm? And if they're not getting firmware from Qualcomm, they're using 3rd-party firmware and that 3rd party would be responsible for licensing unless their contract with Apple says otherwise, in which case it would fall to Apple and Apple wouldn't be so stupid as to take that fight to court where they know they'd lose.
Clearly, if Apple is fighting this, the license fees have been paid somewheresomeone. Qualcomm is too big for Apple to bankrupt with a drawn-out court battle; yes, technically they have the money to do so but, as a publicly traded company, their board of directors would have to answer for the massive dip in profits over however many quarters that took.
At any rate, we're both just talking out our asses right now. Since it looks like Apple intends to take this to court, though, we'll get to see soon enough which of us was right.
Also, real men shoot 45.
Yes, it does address trolls quite well. What it doesn't address is the sole inventor who can only work on his inventions in his spare time because he has bills to pay; how does anything you've proposed thus far protect him from some corporate entity learning what he's working on (maybe he shares a few details with a friend in a bar and someone overhears, for example) and they get the thing built in a week, while he's still months, or a year or more, from finishing?
What you propose would be great for corporations who actually produce the things they invent, but not so great for individuals who may also wish to produce the things they invent, but have other obligations and, so, can't do so as quickly.
Fallout spewed into the upper atmosphere in the US can still reach Russia.
I'd go almost that far, but you must consider the time it might take a single inventor on no budget to build his invention, versus the time it would take a corporate-backed team. I'd say you must actively be building it, and that periodic proof of progress (until completion) should be required in order to have an enforceable patent. This would protect the little guy who might take a year or longer working part-time on building his invention (after all, he has bills to pay) from larger competitors who may otherwise steal his idea by being first to build (after all, they have budgets for R&D, corporate espionage, and manufacturing, as well as staff on hand to work on it full-time in a team capacity).
But, the question is, did they?
If the suppliers stopped paying, Apple should pay licensing for all parts acquired for which the supplier did not. However, if Apple asked suppliers to stop paying and the suppliers did not stop, well, licensing is still being paid and Apple doesn't owe shit.
Well, if Apple is being expected to pay, that's only fair. In fact, Qualcomm should welcome that, as their licensing is based on the final sale price of the finished product; it would be in Qualcomm's best interest to charge their licensing fee as late in the chain as possible.
You do realize that you can now run an full Ubuntu environment within Windows, right? Last I checked, Windows will run Word and Excel natively, too. Cygwin is so 2015.
Yes, one built last month on the AM3+ platform. That's an AMD FX CPU, at least as old as the Mac it's being compared to. Let's see... 8 core... the newest one of those is hmm... the FX-9590, released in 2013.
That's four years old.
Funny, that's the same benchmark you set for beating the three year old Mac.
Because the manufacturers of the chips the use in their phones have already paid those several dollars per unit per patent and passed that cost along to Apple in their unit pricing. In short, Apple has already paid it once, when they bought the chips, and rightly does not feel they should have to pay it again.
This. Especially since the chip manufacturer would already have paid Qualcomm to license their technology, otherwise the chip manufacturer would be the one being sued or Apple would be willing to pay Qualcomm directly as part of their agreement with said chip manufacturer.
Apple continues to use Qualcomm's technology while refusing to pay for it twice
Someone makes the chips Apple uses in their phones. Those chips implement Qualcomm's patented technologies and whoever makes them, therefore, must be licensing those technologies, which means they're paying for them and passing that cost along to Apple who, by paying for the chips which incorporate Qualcomm's technologies, made by the company who already paid for Qualcomm's technologies, has already paid for the use of Qualcomm's technologies.
I realize that's hard for some people (namely Qualcomm's leadership and council) to follow, so let's look at a similar situation with a different type of product.
Coca-Cola uses high fructose corn syrup in their products. They buy this high fructose corn syrup from a supplier. They pay for this high fructose corn syrup when they buy it, then they incorporate it into their products, which they then sell to stores. The stores, then, sell the products to consumers. The stores do not owe the high fructose corn syrup manufacturer anything for the use and sale of their product, because Coca-Cola already paid for that.
Rephrased to fit the Qualcomm situation, with edits bolded to make them clear:
Chip manufacturers use Qualcomm technologies in their products. They license these Qualcomm technologies from a Qualcomm. They pay for these Qualcomm technologies when they license them, then they incorporate them into their products, which they then sell to Apple. Apple, then, sells the products to consumers. Apple does not owe Qualcomm anything for the use and sale of their technologies, because the chip manufacturers already paid for that.
Hahahahahahaha tell me about it, brother. It took me a while to realize you thought I was arguing that prohibition was ineffective...