Trademarks are really only enforceable in marketing material, and even then only if they are misleading as to the origin of the product. Google are quite strict about avoiding mentioning that Android is based on Java in any of their marketing material. Reference material and strictly factual discussion of a product are not subject to trademark restrictions. The fact is, that in order to program for Android, you use Java. Usually, you use Oracle's Java implementation, typically to run Eclipse (although you can choose to use Oracle's dev tools too, if you prefer). The developers reference is therefore bound to mention Java. It would be impossible for them not to.
If I sell a computer, and tell my customer that it doesn't have an operating system with it, but in order to use it they'll need to buy one, and I suggest they buy Microsoft Windows and give them a detailed document describing how to install and use Windows, this isn't a violation of MS's trademarks, despite the fact that I have no license to sell MS's stuff with my computer.
I doubt this will enter into it at all, because that has gone to Jury.
Just to clarify this matter: it is not up to the jury to decide whether there is a valid copyright. The question they are there to answer is, if there is a copyright in Java's API design, did Google infringe it?
If they answer yes (it would be rather perverse if they didn't, but stranger things have happened), then it will be up to the judge to decide on the point of law of whether the copyright is valid or not. If they answer no, he will not have to make any such decision.
The judge has not made a ruling that the copyright is valid; he has told the jury that it is not their responsibility to determine whether it is or not, but that as he is not ready to rule on the question yet they should assume he will answer yes (as this is the answer that allows them to consider whether an infringement occurred, and if he decides there isn't a valid copyright he can simply overrule them later).
OK, you try turning this into a viable commercial product at a lower cost than the competition. The problem is, this is a niche market because these things are really hard to find a suitable application for. You'll be setting up a manufacturing base and then selling maybe 1000 units per year, so you need to offset the cost of manufacturing, support staff, sales staff, development.... hence you'll be selling each unit for $1000 or more. Probably much more, because to make it useful you'll need precision manufacturing (alignment of multiple output beams so they don't diverge over a range of... well, current commercial systems vary between about 500m and 5km, so it won't be trivial). So, no, this isn't a significantly cheaper tech. It's the same tech using similar components that cost about the same (A typical currently available commercial system will use a diode like this one, which might cost 3 times as much as the ones in laser pointers, but it is also capable of 5 times the bandwidth). But it's optical frequency rather than IR, so the range per unit power will be lower, it'll be more disturbed by fog, and it needs multiplexing to reach reasonable data rates, so that will mean more expensive optics at the receiving end. Put this design in a commercial system and you'll see pretty quickly that it's as expensive as existing designs, if not more so.
802.11n uses 250mW transmit power to achieve 600Mbits = about 2.4mW/Mb A laser pointer typically uses about 5mW, so 2 of them will be about 10 mW, for 1Gbit = about 0.01mW/Mb.
But then 802.11n is omnidirectional, whereas a laser is unidirectional, so this is really an apples to htcs comparison.
The reason the available commercial equipment for this stuff is expensive has nothing to do with the quality of the laser, though (the site GP linked to specifies the laser in their entry level device as being a 7mW laser diode, so probably about 50% more powerful than the lasers used in the OP's article). The point is that it's expensive because the only application it's viable for is inter-building linkage, which *almost nobody wants to do*. You can't use it to replace ordinary wireless networks, because it's literally point-to-point: you have to stay stationary in a single pre-determined spot to receive a signal. The only real application is for large companies who have multiple buildings within line of sight of each other. This is a rather unusual situation. Thinking about organisations in my city, there are a couple of universities that could use it, and a hospital. Maybe a few schools. But then these guys can mostly just dig up the land between their buildings and lay cables, which will give higher capacity and more reliability (one of the universities has an issue because one of their buildings is separated from the rest by a public road... they might benefit from this).
No, you can't. Somewhere in the region of 50% of deployed web browsers don't support SNI. Setting up a commercial site that relies on having SNI enabled browsers in order to work correctly would be suicide.
But it does not explain why about 33% of the servers surveyed support SSL v2.0, which virtually no client wants to use, and which is also insecure.
Because, as a server operator, I don't especially care if clients are spoofed. I don't perform any authentication of their identities anyway, so my security doesn't decrease.
If the client wants to use an insecure protocol (or is incorrectly configured to use an insecure protocol in preference to a new one), then that is the client's concern. I'm not going to stop them if they don't want to -- they can turn off SSL2 in their browser options (most modern browsers ship this way anyway) if they care that much. A properly configured browser will use SSL3 or TLS in preference to SSL2 anyway if the server supports it, which mine does, so most people will never notice.
Speaking purely from a commercial standpoint, denying customers access to my services because they are using an out of date or badly configured system makes no sense.
In fact some have said they don't implement it, because most of their players are couch players who aren't interested in using a mouse even if it's supported.
... because the practicalities of the hardware and life virtually force them to be. The hardware's designed to hook up to a TV, the TV is in the living room, most people don't have desks in their living room. It's part of the design decision of how a console works to make it convenient to keep in the area most people spend most time in at the expense of making it easy to use with keyboard/mouse.
Sure, you can get a dedicated TV and use it like a monitor on a desk, but that's a very small minority use, and not what the system designers intended.
Name me the first PC game that could do 1920x1080 at 60fps with 7.1 near-perfect audio...
It's hard to tell, because most PC benchmarks back then used 4:3 resolutions, not 16:9, but given that in 2006 games like FEAR were pushing 140fps on 1024x768, and 1920x1080 is only 2.6 times as many pixels, the same hardware should have managed at least 53fps... and that's assuming that the machine spends all of its time rendering pixels (not likely to be true). I challenge you to tell the difference between 53fps and 60fps without a frame counter.
Not sure it supported 7.1 audio, but it definitely did 5.1, which is good enough.
At ten hours of gaming for $50 or $60, even a short video game is a much better deal than a movie ticket
I get movie tickets for £5 = $8. Typical movie these days runs to about 2hrs, so that's $4/hr, compared to your $5-$6 for a game. Now that's before considering that your $50-$60 for a game isn't realistic pricing for me here -- I'd be looking at £35-45 for a typical new release game, so $57-$73. Your $3/hr for a longer game has become much closer to the $4 for the movie.
Do you really think Amazon care about piracy? All Amazon care about is locking kindle owners in so they can't decide to switch to another manufacturer when someone else has a much better reader available in a couple of years' time.
On the one hand this may not be the precise method Tor is hoping for
I suspect it is. You mention Cory Doctorow later in your post as a proponent of this strategy; note that Tor is his publisher. Tor have experience of the increase in sales he saw when he released free copies of his books. Now note that he is also a friend of one of Tor's most senior editors, who may not set policy but is certainly very influential with those who do, and I think you start to see the picture of how this happened.
Actually, once you have a typeset paper book, typically either in quark or indesign, the cost to convert it to an ebook would be trivial.
Maybe it should be, but that simply doesn't happen a lot of the time. Ebooks are frequently re-typeset, sometimes even retyped (or at least OCR'd and re-copyedited). It's ironic that Stross has come up in this context, because it was when reading one of his Laundry books on Kindle a couple of years ago that I first noticed that it had typos that aren't in the original.
I think, basically, what happens is that the publisher subcontracts the production of the book, and doesn't get the digital files back (I suspect they get a format that isn't readily converted back to a text-like format, such as a list of x/y offsets for characters). It then has to go back to the original source and get somebody else to produce the ebook, requiring a duplication of effort. I imagine some publishers have started moving over to combined producers who do both, but it certainly isn't universal yet.
Second. Baen have been selling DRM-free ebooks for years. In fact, Tor *used* to sell their books through Baen's system, but had to withdraw then, due (I'm led to believe) by pressure from their parent company.
OK, that is evidence of a clean room implementation. How did Sun code end up in Android then?
I see no clear evidence that it did. All the code examples I've seen that are supposedly copied code seem trivial, and likely to be accidental duplication.
He was working for Google in an entirely different capacity, and wrote the code as a translation of a Python implementation. The code was then copied into Android by a second Google employee, who was possibly unaware of its source. But as the code in question is trivial, and represents just about the only sensible way to achieve the same results, I'm not sure it counts as copyright code (see the Lexmark case of a few years ago -- if code is the only sensible way of implementing a functional requirement, no copyright exists). Here is the code in question:
private static void rangeCheck(int arrayLen, int fromIndex, int toIndex) {
if (fromIndex > toIndex)
throw new IllegalArgumentException("fromIndex(" + fromIndex +
") > toIndex(" + toIndex+")");
if (fromIndex < 0)
throw new ArrayIndexOutOfBoundsException(fromIndex);
if (toIndex > arrayLen)
throw new ArrayIndexOutOfBoundsException(toIndex);
}
Note that most restaurants will serve their coffee cooler than 84: that figure only really applies to instant coffee, and they're unlikely to be using instant. Both espresso and filter coffee will likely be substantially cooler. McD, as I understand it, were taking the output of their filter machine and *reheating it* on a hotplate in order to get it up to 88. The reason for doing this was apparently to increase the length of time their staff could wait between refilling the coffee jugs, thus saving a small amount of time in the kitchen.
In 2006, only javac, the java compiler, was open source. Android doesn't even use this compiler, so this was irrelevant to them. It took until 2007 for a GPL release of the class library, and Android was basically finished by this point in time. The first android phone launched only weeks after Java's GPL release. The decision to pursue an open source Java implementation was taken in 2005, shortly after Google acquired Android, and long before Sun began open-sourcing anything.
Isn't this basically what Linus Torvalds did with Linux?
Yes & no. Torvalds was implementing POSIX. There are two advantages he had here:
1. POSIX is defined in a document that is separate from any implementation of it and is published by a group that is not a single copyright holder of a particular implementation. It was specifically intended to be a vendor-neutral standard that anyone could implement. 2. There were already free implementations available (BSD, and the out-of-copyright-due-to-error older versions of System V) that he could use as the basis for his own; both of these had already had their validity argued in court.
Slashdot Mirror
Trademarks are really only enforceable in marketing material, and even then only if they are misleading as to the origin of the product. Google are quite strict about avoiding mentioning that Android is based on Java in any of their marketing material. Reference material and strictly factual discussion of a product are not subject to trademark restrictions. The fact is, that in order to program for Android, you use Java. Usually, you use Oracle's Java implementation, typically to run Eclipse (although you can choose to use Oracle's dev tools too, if you prefer). The developers reference is therefore bound to mention Java. It would be impossible for them not to.
If I sell a computer, and tell my customer that it doesn't have an operating system with it, but in order to use it they'll need to buy one, and I suggest they buy Microsoft Windows and give them a detailed document describing how to install and use Windows, this isn't a violation of MS's trademarks, despite the fact that I have no license to sell MS's stuff with my computer.
I doubt this will enter into it at all, because that has gone to Jury.
Just to clarify this matter: it is not up to the jury to decide whether there is a valid copyright. The question they are there to answer is, if there is a copyright in Java's API design, did Google infringe it?
If they answer yes (it would be rather perverse if they didn't, but stranger things have happened), then it will be up to the judge to decide on the point of law of whether the copyright is valid or not. If they answer no, he will not have to make any such decision.
The judge has not made a ruling that the copyright is valid; he has told the jury that it is not their responsibility to determine whether it is or not, but that as he is not ready to rule on the question yet they should assume he will answer yes (as this is the answer that allows them to consider whether an infringement occurred, and if he decides there isn't a valid copyright he can simply overrule them later).
That made me think of blue lasers, which would have even better rates.
But, how about longer waves, such as infrared or even radio?
A typical 1.5mW near-IR laser diode can emit at 2.5Gb/s and costs only about 3 times as much as a laser pointer, so is more economical.
OK, you try turning this into a viable commercial product at a lower cost than the competition. The problem is, this is a niche market because these things are really hard to find a suitable application for. You'll be setting up a manufacturing base and then selling maybe 1000 units per year, so you need to offset the cost of manufacturing, support staff, sales staff, development.... hence you'll be selling each unit for $1000 or more. Probably much more, because to make it useful you'll need precision manufacturing (alignment of multiple output beams so they don't diverge over a range of... well, current commercial systems vary between about 500m and 5km, so it won't be trivial). So, no, this isn't a significantly cheaper tech. It's the same tech using similar components that cost about the same (A typical currently available commercial system will use a diode like this one, which might cost 3 times as much as the ones in laser pointers, but it is also capable of 5 times the bandwidth). But it's optical frequency rather than IR, so the range per unit power will be lower, it'll be more disturbed by fog, and it needs multiplexing to reach reasonable data rates, so that will mean more expensive optics at the receiving end. Put this design in a commercial system and you'll see pretty quickly that it's as expensive as existing designs, if not more so.
802.11n uses 250mW transmit power to achieve 600Mbits = about 2.4mW/Mb
A laser pointer typically uses about 5mW, so 2 of them will be about 10 mW, for 1Gbit = about 0.01mW/Mb.
But then 802.11n is omnidirectional, whereas a laser is unidirectional, so this is really an apples to htcs comparison.
The reason the available commercial equipment for this stuff is expensive has nothing to do with the quality of the laser, though (the site GP linked to specifies the laser in their entry level device as being a 7mW laser diode, so probably about 50% more powerful than the lasers used in the OP's article). The point is that it's expensive because the only application it's viable for is inter-building linkage, which *almost nobody wants to do*. You can't use it to replace ordinary wireless networks, because it's literally point-to-point: you have to stay stationary in a single pre-determined spot to receive a signal. The only real application is for large companies who have multiple buildings within line of sight of each other. This is a rather unusual situation. Thinking about organisations in my city, there are a couple of universities that could use it, and a hospital. Maybe a few schools. But then these guys can mostly just dig up the land between their buildings and lay cables, which will give higher capacity and more reliability (one of the universities has an issue because one of their buildings is separated from the rest by a public road... they might benefit from this).
You can't?
No, you can't. Somewhere in the region of 50% of deployed web browsers don't support SNI. Setting up a commercial site that relies on having SNI enabled browsers in order to work correctly would be suicide.
But it does not explain why about 33% of the servers surveyed support SSL v2.0, which virtually no client wants to use, and which is also insecure.
Because, as a server operator, I don't especially care if clients are spoofed. I don't perform any authentication of their identities anyway, so my security doesn't decrease.
If the client wants to use an insecure protocol (or is incorrectly configured to use an insecure protocol in preference to a new one), then that is the client's concern. I'm not going to stop them if they don't want to -- they can turn off SSL2 in their browser options (most modern browsers ship this way anyway) if they care that much. A properly configured browser will use SSL3 or TLS in preference to SSL2 anyway if the server supports it, which mine does, so most people will never notice.
Speaking purely from a commercial standpoint, denying customers access to my services because they are using an out of date or badly configured system makes no sense.
In fact some have said they don't implement it, because most of their players are couch players who aren't interested in using a mouse even if it's supported.
... because the practicalities of the hardware and life virtually force them to be. The hardware's designed to hook up to a TV, the TV is in the living room, most people don't have desks in their living room. It's part of the design decision of how a console works to make it convenient to keep in the area most people spend most time in at the expense of making it easy to use with keyboard/mouse.
Sure, you can get a dedicated TV and use it like a monitor on a desk, but that's a very small minority use, and not what the system designers intended.
Name me the first PC game that could do 1920x1080 at 60fps with 7.1 near-perfect audio ...
It's hard to tell, because most PC benchmarks back then used 4:3 resolutions, not 16:9, but given that in 2006 games like FEAR were pushing 140fps on 1024x768, and 1920x1080 is only 2.6 times as many pixels, the same hardware should have managed at least 53fps... and that's assuming that the machine spends all of its time rendering pixels (not likely to be true). I challenge you to tell the difference between 53fps and 60fps without a frame counter.
Not sure it supported 7.1 audio, but it definitely did 5.1, which is good enough.
At ten hours of gaming for $50 or $60, even a short video game is a much better deal than a movie ticket
I get movie tickets for £5 = $8. Typical movie these days runs to about 2hrs, so that's $4/hr, compared to your $5-$6 for a game. Now that's before considering that your $50-$60 for a game isn't realistic pricing for me here -- I'd be looking at £35-45 for a typical new release game, so $57-$73. Your $3/hr for a longer game has become much closer to the $4 for the movie.
If the next Call of Duty game had shoddy graphics to save money, can you imagine the reaction?
If it cost half as much, I'd probably buy it.
Do you really think Amazon care about piracy? All Amazon care about is locking kindle owners in so they can't decide to switch to another manufacturer when someone else has a much better reader available in a couple of years' time.
On the one hand this may not be the precise method Tor is hoping for
I suspect it is. You mention Cory Doctorow later in your post as a proponent of this strategy; note that Tor is his publisher. Tor have experience of the increase in sales he saw when he released free copies of his books. Now note that he is also a friend of one of Tor's most senior editors, who may not set policy but is certainly very influential with those who do, and I think you start to see the picture of how this happened.
Actually, once you have a typeset paper book, typically either in quark or indesign, the cost to convert it to an ebook would be trivial.
Maybe it should be, but that simply doesn't happen a lot of the time. Ebooks are frequently re-typeset, sometimes even retyped (or at least OCR'd and re-copyedited). It's ironic that Stross has come up in this context, because it was when reading one of his Laundry books on Kindle a couple of years ago that I first noticed that it had typos that aren't in the original.
I think, basically, what happens is that the publisher subcontracts the production of the book, and doesn't get the digital files back (I suspect they get a format that isn't readily converted back to a text-like format, such as a list of x/y offsets for characters). It then has to go back to the original source and get somebody else to produce the ebook, requiring a duplication of effort. I imagine some publishers have started moving over to combined producers who do both, but it certainly isn't universal yet.
Second. Baen have been selling DRM-free ebooks for years. In fact, Tor *used* to sell their books through Baen's system, but had to withdraw then, due (I'm led to believe) by pressure from their parent company.
http://code.google.com/p/sacd-ripper/
http://forum.doom9.org/showthread.php?t=96860
Found within moments of googling for each.
803(6)?
You Forth about talking are, I think is what you're aiming for. Your sentence came across as more German than RPN.
"Logically consistent" and "able to be used to prove its own consistency" are not the same thing.
OK, that is evidence of a clean room implementation. How did Sun code end up in Android then?
I see no clear evidence that it did. All the code examples I've seen that are supposedly copied code seem trivial, and likely to be accidental duplication.
Also, what was this guy doing working on Android?
He wasn't. http://www.zdnet.com.au/google-oracle-get-technical-in-court-339336340.htm
He was working for Google in an entirely different capacity, and wrote the code as a translation of a Python implementation. The code was then copied into Android by a second Google employee, who was possibly unaware of its source. But as the code in question is trivial, and represents just about the only sensible way to achieve the same results, I'm not sure it counts as copyright code (see the Lexmark case of a few years ago -- if code is the only sensible way of implementing a functional requirement, no copyright exists). Here is the code in question:
Sure looks trivial to me.
Note that most restaurants will serve their coffee cooler than 84: that figure only really applies to instant coffee, and they're unlikely to be using instant. Both espresso and filter coffee will likely be substantially cooler. McD, as I understand it, were taking the output of their filter machine and *reheating it* on a hotplate in order to get it up to 88. The reason for doing this was apparently to increase the length of time their staff could wait between refilling the coffee jugs, thus saving a small amount of time in the kitchen.
In 2006, only javac, the java compiler, was open source. Android doesn't even use this compiler, so this was irrelevant to them. It took until 2007 for a GPL release of the class library, and Android was basically finished by this point in time. The first android phone launched only weeks after Java's GPL release. The decision to pursue an open source Java implementation was taken in 2005, shortly after Google acquired Android, and long before Sun began open-sourcing anything.
I did. All I get is a series of articles that suggest he probably didn't "steal" anything.
Isn't this basically what Linus Torvalds did with Linux?
Yes & no. Torvalds was implementing POSIX. There are two advantages he had here:
1. POSIX is defined in a document that is separate from any implementation of it and is published by a group that is not a single copyright holder of a particular implementation. It was specifically intended to be a vendor-neutral standard that anyone could implement.
2. There were already free implementations available (BSD, and the out-of-copyright-due-to-error older versions of System V) that he could use as the basis for his own; both of these had already had their validity argued in court.