If i give you something for free and know that you would be breaking the law if you actually used then that is a pretty immoral thing to do.
In as much as infringing on patents is "breaking the law", everybody does it, just to keep doing business. So every tech company is immoral in this respect, and very nearly equally so.
Since the other person also knows about patents, he is expected to do a research and accept the free gift only if they have patents in their portfolios to counter-sue the litigators if they pop up. Nothing immoral here in this respect. It would have been immoral if Google promised that the free does not have patent risks, or that Google will protect the licensees from patent litigators but then actually not defend.
Do you really think Motorola, HTC, Samsung and friends do not have the ability to research for patents? That they didn't know Android could be a patent minefield? Being based on linux itself must have sent warning bells ringing, given Microsoft's rhetoric about the secret patents. They are no kids who were handed poisoned toffee by Google.
It's to be expected if something similar happened in Calcutta, the outcome would be even worse
Something much worse happened in Calcutta's state (West Bengal). It was followed up by other acts of state terrorism. People of West Bengal reacted by democratically and peacefully ousting a 35 year old monopoly of the left front in West Bengal. Yes, it took 3 years, but at least they didn't react by looting shops.
I am no fan of people of Calcutta, or even West Bengal, but they deserve much more respect than you are giving them.
I've got McDonald friends who've never heard the bagpipes, and Russos who've never cooked lasagne
Then learn about large scale data mining. Hint - exceptions are far from fatal.
Much easier in that case to target people who've joined Spanish groups, who list Spanish as a language they speak, or identify themselves as Spanish
It works both ways. Why have they joined Spanish groups? Because the networking sites suggested Spanish groups to them based on a lot of data - including surname. You dismiss surname completely as a way to estimate ethnicity probability - I say that it is wrong to dismiss.
Zip code? Read my comment - I said that zip code is useful for advertising...... door-knocking a zipcode trying to track down the target is likely to lead to very tired and frustrated would-be stalker.
I replied to the specific part I quoted - that you never heard of an advertizer "wanting to target only people with a surname of Rodriguez, or who live in 89th street; city or zip code is as much as they really care about.". So I am not really replying to the stalker argument, only your ignorance about advertizer behaviour.
Zip code helps, but very frequently a place is close to 3 different zip codes which are not even adjacent numerically, but the place is on 89th street. Proprietors owning such place would rather buy an adword for "city + locality + 89th street" rather than the zip code. Yes, zip code adwords are more expensive and in general more valuable, but in specific cases, street number adwords can be better.
In large scale data mining, every bit helps - you combine all information that you can get and come up with a correlation coefficient. Street, surname might not be perfect but it is highly ignorant to suggest they are worthless data points.
I've never heard of an advertiser wanting to target only people with a surname of Rodriguez, or who live in 89th street; city or zip code is as much as they really care about.
You have not heard much, then, have you?
Rodriguez : Spanish clubs / restaurants / community centres would pay much more to advertize to a Rodriguez than to a Churchill.
89th street / zip code : Starbucks franchisee of that area? A new gym in that area?
Not exactly. When cost of not replacing hardware (in increased power consumption, rack space, cooling requirement etc) exceeds the cost of replacing hardware ; they replace. Even when the trashed hardware doesn't "actually break".
AMD APU, which has some really wicked features like letting the integrated do physics while the discrete takes care of textures and by having such tight integration from the looks of it for FP the AMD chips are gonna rock..
While the APU idea is great, current implementations are neither here nor here :
1. GPU portion too slow for gamers, overkill for non-gamers. Sandy Bridge integrated graphics is enough for non-gamers, gamers have to get a dedicated graphics card still.
2. Non-gamer might still get it to be "future proof", but then CPU portion is too slow for "future proof". It is great for most tasks, but not really future proof material.
3. APU works with a dedicated AMD graphics card, but a huge variety of 4 current graphics cards and none to come in the future are supported by APU. So not exactly ideal for an aspiring gamer hoping to upgrade later either.
Hope next series from AMD strikes a better balance.
But birds make music all the same, whatever the purpose. Do you think humans make music to make music? They make music to make money, sex, enjoyment. Same as with birds (except for money)
Is reading before replying out of fashion these days?
Your initial 3 links are about US market, whereas TFA (and discussion until you butted in) is about global market.
Your 4th link is idiotic - the reports are not comparing an OS and a phone. They are comparing "mobile phones running iOS" and "mobile phones running Android". Since only Apple makes the former, it is colloquially said as iPhones vs. Android phones. Not to mention, after saying this, the author himself goes and compares iPhone + iPod touch + iPad to Android (which is mostly selling on phones) - which is useless because of comparing devices of so different sizes that there is practically no overlap between customer base.
In tablets, iPad is doing much better than all Android tablets put together, almost everywhere.
You will do well to read the monkey-banana-cage-water story. Religiously biased are like those monkeys.
The religious bias might have an advantage, or might have had an advantage, or could be random. If you follow them simply believe them because they are religious, you will not tend to analyze whether they are still relevant, or even if they were ever relevant.
You think birds or whales sing to make music? I doubt any species but us make music.
But birds make music all the same, whatever the purpose. Do you think humans make music to make music? They make music to make money, sex, enjoyment. Same as with birds (except for money)
Microsoft's only fault in the case had come out to be that they had altered the behaviour of java.* packages and org.sun.* packages. Which Google hasn't. Microsoft should have included their "extensions" in org.microsoft.* packages.
Those days Sun chose mostly not to sue on java patents on simply implementing another JVM as long as it was standard compliant and hence not vulnerable to Extend Embrace Extinguish attack.
Ok, point taken. I got misled by an apparent contradiction in your post.
But the US HTC Desires (forgot the carrier) that I have seen, even unlocked and unrooted, allow the "wifi hotspot". Is it not true for all carriers?
Not sure about tethering directly, though the wifi hotspot is even more convenient to "steal" bandwidth these days with laptops replacing most desktops - and laptops typically have wifi receiver.
On the other hand, iOS doesn't have the hotspot feature (last I checked). Even in countries where it is sold unlocked. Since iPhone hardware supports this, jailbroken iPhones can install an application for the feature.
this is a fee your CARRIER imposes and has nothing to do with the phone (or its jailbrokenness).....blah blah... Jailbreaking gets around that because
Have you any sense of logic? Read what you wrote above - If jailbreaking gets around carrier fee, how does it have nothing to do with the carrier fee?
Bypass surgery gets around blocked blood vessels, but has nothing to do with blocked blood vessels? I am shocked at the the level of insanity you fanboys need to cultivate!
It was along the lines of "come on, what reasons are there really to jailbreak these days anyway?".
Actually, it was "Name one legitimate reason to want to jailbreak your phone now days."
what relevancy do you think this could possibly have to the market in general?
Who said anything about the market in general? In case you didn't notice, this is Slashdot. News for nerds. You address a nerd and ask a "reason to want to jailbreak your phone". Guess what? You are going to get a nerd reply. Big surprise.
Then you come, and try to find relevancy to "market in general". And by your own admission and endless repetition, Slashdotters have nothing to do with "market in general". So, your response is irrelevant. Got it?
For a thinking person, the knee jerk reactions are likely to balance out over time. Sometimes they would be in favour of the company (or anything one can be a fanboy of) and sometimes against. Still I see some Slashdot posters who always have positives to speak about some companies. I am hard pressed to conclude anything other than fanboy-syndrome or shill employment.
So, you're saying that you can institute an arbitrary scheduling algorithm with a config value change, even if the code for the new algorithm isn't part of the system?
Not sure what you mean by "the system". For the linux example I was talking about, the kernel is compiled with multiple scheduling algorithms and the config values switch between them. Some algorithms themselves are parametrized, so config values also specify the parameters.
they tend to need to change a lot more than just NTOSKRNL.EXE for such updates, so it's either a major service pack or a completely new version.
It is completely legal to change the "lot more than just NTOSKRNL.EXE" that is required, even with a free update. It is also not illegal to name the update anything other than "major service pack". Likewise, it is totally constitutional to supply these scheduling algorithm changes without charging for it.
So it comes down to a business decision, calculated to maximize profits. Which is what I said. Your response is irrelevant - doesn't matter which EXE and which $SYS$ file is changed.
Most of the time on Slashdot people are labeled 'fanboys' when they actually come up with a good argument and a good rebuttal is not immediately forthcoming. This is often followed by an accusation of being on said company's payroll.
That can happen. Though do you think it is conceivable that a person impartially and logically always comes to a positive conclusion about a given company? Across hundreds of slashdot posts - regarding the share price of the company, product usability, company ethics, product longevity, environmental impact, nobility of the CEO ? A belief in the impeccability of the company charter as well as well the company's adherence to it over decades?
As General Patton put it so well : If everyone is thinking alike, then somebody isn't thinking.
You are still thinking raw CPU power still matters. In a world where even web browsers are 3D accelerated, the GPU suddenly becomes extremely important, even more than the CPU
Intel integrated graphics is enough for Aero, 99% of the flash videos and games in the internet, and more.
If Bobcat and Llano are any indication, AMD will integrate a GPU that will be at least 2-4 times faster than the GPU in Sandy Bridge while consuming the same amount of power. And if some of the reviews I read are correct, the integrated AMD GPU will be able to work together with the discrete GPU for a 30-70% performance boost.
There is an extremely limited number of discrete GPUs the APUs can work together with. For Llano, it's just 4 discrete GPUs. And these 4 are not even the fastest of today.
Now if someone buys a high end system with a discrete GPU, Bobcat will still be faster, because the integrated GPU will work with the discrete GPU
The consumer market can be divided into the following segments :
1. Hardcore Gamers : Sandy/Ivy bridge are better. Gamers will never settle with integrated graphics of Llano (even though it is 2-4 times better than Intel integrated graphics). Nor will they use the 4 GPUs Llano works with. They will also upgrade their GPUs at thrice the frequency of upgrading the CPUs - and all the upgraded GPU will be unable to work with their APUs. Better go with Intel, simply not use the integrated graphics but get the fastest GPU they can afford and keep upgrading as frequently as possible. This is 1% of the market but the profitable part of the market.
2. Web / facebook / email / MS word : Sandy/Ivy Bridge integrated graphics is and will be more than adequate. In fact, Llano's integrated graphics is an overkill for such things. In return, they can get faster CPU for occasional CPU intensive tasks. It is years before mainstream software will start using GPU power for general computing. This is 95% of the market, though not much profit per consumer here.
3. Specialized consumer : Uses specialized professional software or unusual in other respects. Such consumer may or may not prefer Llano - depending on their multi-tasking requirement. Llano excels at multi-tasking, but very few people really work on more than one task at a time - and when they do work on it rarely they can see some delay there to see benefit in their 99% workload.
Why does a new scheduler need a new OS release? I have for years changed scheduler by a simple reboot (in linux, of course). I hear it can be changed by an even simpler echo into some system file.
Just illustrating that nothing needs a new OS. It's just a business decision about which features to club together in a new OS to give an impression of being nice to their customers while making good profits. With Microsoft, their (near) monopoly is also considered while making this decision.
So what is the problem? Driving a car and drinking from a straw are not "tool use (TM)" then. Once something is defined in a certain way, it is worth it to stick to that definition. Why must your every activity classify as tool-use under every definition to boost your fragile self-worth?
Aren't you capable of sharpening your pencil when it loses its point? Tool use. Filling gas in your car's gas tank to be able to drive it? Tool use. Uncorking a bottle to drink using a cork-screw / bottle opener? Tool use. Installing an operating system to use your "One True OS (TM)" ? Tool use (yeah, non-slashdotters fail in this particular instance).
as long as both parties are happy with it and are honest about it? If they are not, then that is probably as a result of social memes not because of honesty
You just contradicted yourself.
Anyway, if you trust X not to do Y, then X does Y anyway, it is still a breach of trust. For any value of Y, you can argue that what has trust got to do with Y, but you would be wrong. Because trust does not stand alone, trust is on a person to do or not to do certain things, in this case not to do Y.
It is like I promise to give someone an iPhone for 200$ and the other person trusts me to fulfil this promise. Now I refuse to give the iPhone. He says it is breach of trust because he trusted me to give him an iPhone for 200$. I say "What has trust got to do with iPhone?". Stupid, isn't it? So is your post.
That's right, small companies / individual software developers have a huge opportunity in the form of app stores and markets. Kudos to Apple for starting that, and Google to follow the example.
Slashdot Mirror
If i give you something for free and know that you would be breaking the law if you actually used then that is a pretty immoral thing to do.
In as much as infringing on patents is "breaking the law", everybody does it, just to keep doing business. So every tech company is immoral in this respect, and very nearly equally so.
Since the other person also knows about patents, he is expected to do a research and accept the free gift only if they have patents in their portfolios to counter-sue the litigators if they pop up. Nothing immoral here in this respect. It would have been immoral if Google promised that the free does not have patent risks, or that Google will protect the licensees from patent litigators but then actually not defend.
Do you really think Motorola, HTC, Samsung and friends do not have the ability to research for patents? That they didn't know Android could be a patent minefield? Being based on linux itself must have sent warning bells ringing, given Microsoft's rhetoric about the secret patents. They are no kids who were handed poisoned toffee by Google.
It's to be expected if something similar happened in Calcutta, the outcome would be even worse
Something much worse happened in Calcutta's state (West Bengal). It was followed up by other acts of state terrorism. People of West Bengal reacted by democratically and peacefully ousting a 35 year old monopoly of the left front in West Bengal. Yes, it took 3 years, but at least they didn't react by looting shops.
I am no fan of people of Calcutta, or even West Bengal, but they deserve much more respect than you are giving them.
I've got McDonald friends who've never heard the bagpipes, and Russos who've never cooked lasagne
Then learn about large scale data mining. Hint - exceptions are far from fatal.
Much easier in that case to target people who've joined Spanish groups, who list Spanish as a language they speak, or identify themselves as Spanish
It works both ways. Why have they joined Spanish groups? Because the networking sites suggested Spanish groups to them based on a lot of data - including surname. You dismiss surname completely as a way to estimate ethnicity probability - I say that it is wrong to dismiss.
Zip code? Read my comment - I said that zip code is useful for advertising. ..... door-knocking a zipcode trying to track down the target is likely to lead to very tired and frustrated would-be stalker.
I replied to the specific part I quoted - that you never heard of an advertizer "wanting to target only people with a surname of Rodriguez, or who live in 89th street; city or zip code is as much as they really care about.". So I am not really replying to the stalker argument, only your ignorance about advertizer behaviour.
Zip code helps, but very frequently a place is close to 3 different zip codes which are not even adjacent numerically, but the place is on 89th street. Proprietors owning such place would rather buy an adword for "city + locality + 89th street" rather than the zip code. Yes, zip code adwords are more expensive and in general more valuable, but in specific cases, street number adwords can be better.
In large scale data mining, every bit helps - you combine all information that you can get and come up with a correlation coefficient. Street, surname might not be perfect but it is highly ignorant to suggest they are worthless data points.
I've never heard of an advertiser wanting to target only people with a surname of Rodriguez, or who live in 89th street; city or zip code is as much as they really care about.
You have not heard much, then, have you?
Rodriguez : Spanish clubs / restaurants / community centres would pay much more to advertize to a Rodriguez than to a Churchill.
89th street / zip code : Starbucks franchisee of that area? A new gym in that area?
Not exactly. When cost of not replacing hardware (in increased power consumption, rack space, cooling requirement etc) exceeds the cost of replacing hardware ; they replace. Even when the trashed hardware doesn't "actually break".
AMD APU, which has some really wicked features like letting the integrated do physics while the discrete takes care of textures and by having such tight integration from the looks of it for FP the AMD chips are gonna rock..
While the APU idea is great, current implementations are neither here nor here :
1. GPU portion too slow for gamers, overkill for non-gamers. Sandy Bridge integrated graphics is enough for non-gamers, gamers have to get a dedicated graphics card still.
2. Non-gamer might still get it to be "future proof", but then CPU portion is too slow for "future proof". It is great for most tasks, but not really future proof material.
3. APU works with a dedicated AMD graphics card, but a huge variety of 4 current graphics cards and none to come in the future are supported by APU. So not exactly ideal for an aspiring gamer hoping to upgrade later either.
Hope next series from AMD strikes a better balance.
But birds make music all the same, whatever the purpose. Do you think humans make music to make music? They make music to make money, sex, enjoyment. Same as with birds (except for money)
Is reading before replying out of fashion these days?
Your initial 3 links are about US market, whereas TFA (and discussion until you butted in) is about global market.
Your 4th link is idiotic - the reports are not comparing an OS and a phone. They are comparing "mobile phones running iOS" and "mobile phones running Android". Since only Apple makes the former, it is colloquially said as iPhones vs. Android phones. Not to mention, after saying this, the author himself goes and compares iPhone + iPod touch + iPad to Android (which is mostly selling on phones) - which is useless because of comparing devices of so different sizes that there is practically no overlap between customer base.
In tablets, iPad is doing much better than all Android tablets put together, almost everywhere.
You will do well to read the monkey-banana-cage-water story. Religiously biased are like those monkeys.
The religious bias might have an advantage, or might have had an advantage, or could be random. If you follow them simply believe them because they are religious, you will not tend to analyze whether they are still relevant, or even if they were ever relevant.
You think birds or whales sing to make music? I doubt any species but us make music.
But birds make music all the same, whatever the purpose. Do you think humans make music to make music? They make music to make money, sex, enjoyment. Same as with birds (except for money)
That would be python communities.
Microsoft's only fault in the case had come out to be that they had altered the behaviour of java.* packages and org.sun.* packages. Which Google hasn't. Microsoft should have included their "extensions" in org.microsoft.* packages.
Those days Sun chose mostly not to sue on java patents on simply implementing another JVM as long as it was standard compliant and hence not vulnerable to Extend Embrace Extinguish attack.
Ok, point taken. I got misled by an apparent contradiction in your post.
But the US HTC Desires (forgot the carrier) that I have seen, even unlocked and unrooted, allow the "wifi hotspot". Is it not true for all carriers?
Not sure about tethering directly, though the wifi hotspot is even more convenient to "steal" bandwidth these days with laptops replacing most desktops - and laptops typically have wifi receiver.
On the other hand, iOS doesn't have the hotspot feature (last I checked). Even in countries where it is sold unlocked. Since iPhone hardware supports this, jailbroken iPhones can install an application for the feature.
A question. If phone runs out of battery, would this also lose the jailbreak on the tethered jailbreak?
I see that phones don't run out of battery very frequently, but still, nice to know.
this is a fee your CARRIER imposes and has nothing to do with the phone (or its jailbrokenness). ....blah blah ... Jailbreaking gets around that because
Have you any sense of logic? Read what you wrote above - If jailbreaking gets around carrier fee, how does it have nothing to do with the carrier fee?
Bypass surgery gets around blocked blood vessels, but has nothing to do with blocked blood vessels? I am shocked at the the level of insanity you fanboys need to cultivate!
It was along the lines of "come on, what reasons are there really to jailbreak these days anyway?".
Actually, it was "Name one legitimate reason to want to jailbreak your phone now days."
what relevancy do you think this could possibly have to the market in general?
Who said anything about the market in general? In case you didn't notice, this is Slashdot. News for nerds. You address a nerd and ask a "reason to want to jailbreak your phone". Guess what? You are going to get a nerd reply. Big surprise.
Then you come, and try to find relevancy to "market in general". And by your own admission and endless repetition, Slashdotters have nothing to do with "market in general". So, your response is irrelevant. Got it?
For a thinking person, the knee jerk reactions are likely to balance out over time. Sometimes they would be in favour of the company (or anything one can be a fanboy of) and sometimes against. Still I see some Slashdot posters who always have positives to speak about some companies. I am hard pressed to conclude anything other than fanboy-syndrome or shill employment.
So, you're saying that you can institute an arbitrary scheduling algorithm with a config value change, even if the code for the new algorithm isn't part of the system?
Not sure what you mean by "the system". For the linux example I was talking about, the kernel is compiled with multiple scheduling algorithms and the config values switch between them. Some algorithms themselves are parametrized, so config values also specify the parameters.
they tend to need to change a lot more than just NTOSKRNL.EXE for such updates, so it's either a major service pack or a completely new version.
It is completely legal to change the "lot more than just NTOSKRNL.EXE" that is required, even with a free update. It is also not illegal to name the update anything other than "major service pack". Likewise, it is totally constitutional to supply these scheduling algorithm changes without charging for it.
So it comes down to a business decision, calculated to maximize profits. Which is what I said. Your response is irrelevant - doesn't matter which EXE and which $SYS$ file is changed.
Most of the time on Slashdot people are labeled 'fanboys' when they actually come up with a good argument and a good rebuttal is not immediately forthcoming. This is often followed by an accusation of being on said company's payroll.
That can happen. Though do you think it is conceivable that a person impartially and logically always comes to a positive conclusion about a given company? Across hundreds of slashdot posts - regarding the share price of the company, product usability, company ethics, product longevity, environmental impact, nobility of the CEO ? A belief in the impeccability of the company charter as well as well the company's adherence to it over decades?
As General Patton put it so well : If everyone is thinking alike, then somebody isn't thinking.
That is when I positively conclude fanboyism.
You are still thinking raw CPU power still matters. In a world where even web browsers are 3D accelerated, the GPU suddenly becomes extremely important, even more than the CPU
Intel integrated graphics is enough for Aero, 99% of the flash videos and games in the internet, and more.
If Bobcat and Llano are any indication, AMD will integrate a GPU that will be at least 2-4 times faster than the GPU in Sandy Bridge while consuming the same amount of power. And if some of the reviews I read are correct, the integrated AMD GPU will be able to work together with the discrete GPU for a 30-70% performance boost.
There is an extremely limited number of discrete GPUs the APUs can work together with. For Llano, it's just 4 discrete GPUs. And these 4 are not even the fastest of today.
Now if someone buys a high end system with a discrete GPU, Bobcat will still be faster, because the integrated GPU will work with the discrete GPU
The consumer market can be divided into the following segments :
1. Hardcore Gamers : Sandy/Ivy bridge are better. Gamers will never settle with integrated graphics of Llano (even though it is 2-4 times better than Intel integrated graphics). Nor will they use the 4 GPUs Llano works with. They will also upgrade their GPUs at thrice the frequency of upgrading the CPUs - and all the upgraded GPU will be unable to work with their APUs. Better go with Intel, simply not use the integrated graphics but get the fastest GPU they can afford and keep upgrading as frequently as possible. This is 1% of the market but the profitable part of the market.
2. Web / facebook / email / MS word : Sandy/Ivy Bridge integrated graphics is and will be more than adequate. In fact, Llano's integrated graphics is an overkill for such things. In return, they can get faster CPU for occasional CPU intensive tasks. It is years before mainstream software will start using GPU power for general computing. This is 95% of the market, though not much profit per consumer here.
3. Specialized consumer : Uses specialized professional software or unusual in other respects. Such consumer may or may not prefer Llano - depending on their multi-tasking requirement. Llano excels at multi-tasking, but very few people really work on more than one task at a time - and when they do work on it rarely they can see some delay there to see benefit in their 99% workload.
Why does a new scheduler need a new OS release? I have for years changed scheduler by a simple reboot (in linux, of course). I hear it can be changed by an even simpler echo into some system file.
Just illustrating that nothing needs a new OS. It's just a business decision about which features to club together in a new OS to give an impression of being nice to their customers while making good profits. With Microsoft, their (near) monopoly is also considered while making this decision.
So what is the problem? Driving a car and drinking from a straw are not "tool use (TM)" then. Once something is defined in a certain way, it is worth it to stick to that definition. Why must your every activity classify as tool-use under every definition to boost your fragile self-worth?
Aren't you capable of sharpening your pencil when it loses its point? Tool use. Filling gas in your car's gas tank to be able to drive it? Tool use. Uncorking a bottle to drink using a cork-screw / bottle opener? Tool use. Installing an operating system to use your "One True OS (TM)" ? Tool use (yeah, non-slashdotters fail in this particular instance).
as long as both parties are happy with it and are honest about it? If they are not, then that is probably as a result of social memes not because of honesty
You just contradicted yourself.
Anyway, if you trust X not to do Y, then X does Y anyway, it is still a breach of trust. For any value of Y, you can argue that what has trust got to do with Y, but you would be wrong. Because trust does not stand alone, trust is on a person to do or not to do certain things, in this case not to do Y.
It is like I promise to give someone an iPhone for 200$ and the other person trusts me to fulfil this promise. Now I refuse to give the iPhone. He says it is breach of trust because he trusted me to give him an iPhone for 200$. I say "What has trust got to do with iPhone?". Stupid, isn't it? So is your post.
That's right, small companies / individual software developers have a huge opportunity in the form of app stores and markets. Kudos to Apple for starting that, and Google to follow the example.
Now Zynga's a multibillion dollar company, and there's tons more companies making a killing on mobile apps.
While you are correct that Apple did a lot of things right with iPhone, but ...
Zynga owes its billions to Facebook rather than Apple; by a huge extent. There is no iPhone app company worth billions.