Google have implemented 1 and 2 according to the specs.
No, Android does not implement the JVM. Android's Dalvik VM is a register based machine, just like the CLR. In part, that's probably because they wanted to minimize the risk of getting successfully sued by Sun. Let's hope they did their homework.
Not even Sun/Oracle is insane enough to propose that you should use the java standard edition on mobile phones.
No, instead they want you to license their embedded Java implementation, which isn't covered even by the hokey J2SE specs/licenses.
It's also your and my problem if we rely on Java. That's why we shouldn't rely on Java.
Then why are there so many FOSS implementations?
Because Sun/Oracle hasn't bothered suing them out of existence yet. With its patent portfolio, Oracle has the power to kill pretty much any unlicensed implementation of Java.
Java is a spec and NOT an implementation. You are free to make your own implementation of the spec (IBM, Apple and many others do).
IBM and Apple have not "made their own implementations"; they have licensed Sun/Oracle's implementations and created derivatives.
You are not free to make your own implementation of the spec; you need to pass Sun's compatibility tests if you don't want to get hit by patent lawsuits because Sun hold essential patents for creating a conforming implementation.
For years now, there has been no implementation of Java conforming to the Java spec except for those derived from Sun's source code. That's not an accident: it's pretty much impossible to meet Sun's compatibility requirements without licensing their source code.
This would not be true if Apple cheated and did not implement some part of Java spec (which is the case with what Google did).
Google didn't "cheat", Google implemented their own platform and runtime; they just happened to use the Java language to do it. In principle, Sun/Oracle couldn't have done anything about that: Sun doesn't hold a patent on the Java language itself. But it appears as if the Android designers may not have been careful enough to avoid all of Sun's patents.
No, this isn't "like any other company" or like any other patent portfolio. Sun claimed that Java was an open platform when, in fact, it was highly patent encumbered, by design.
And open source developers had other choices. They chose Java because they erroneously believed that Sun's platform was open.
My comment above was a comment about Java, not Android.
What you quote about Android is correct in that it isn't an implementation of the Java platform. However, the patents that Oracle now owns do apply to Android if Android uses the techniques described in those patents (which I believe it does).
Of course it's intentional. It means that Sun effectively retained full control over all Java implementations, because the only implementations that could meet those criteria were the ones passing their compatibility tests and (in practice) only those based on their licensed source code. That makes Java a highly proprietary platform.
And where is your "run anywhere" going to go if Oracle decides not to renew Java licenses to Apple or IBM anymore? I don't think Apple or IBM can use the GPL version of Java on their platforms because they need to link non-GPL code into the implementation.
Java is, and always has been, a proprietary platform, in the same sense as Microsoft Windows has been. All Sun wanted to do was insert their proprietary layer in between the operating system and applications, meaning that you would be dependent on them instead of Microsoft. Looks like they may have the legal means to make it all stick.
"Java" isn't licensed under the GPL. A single Java implementation, derived from Sun's proprietary source code, is licensed under the GPL. Furthermore, the patent grant applies only if you meet specific compatibility conditions, which no implementation other than Sun's meets.
Google implemented the Java language, not its libraries, and did it by themselves. Android (and Dalvik) are licensed under a mix of Apache and GPL, but that doesn't matter; the license under which a third party implementation is released is not relevant for the patent grant.
Google rolled their own implementation and libraries for good reason: the full Java platform would have been far too obese for Android, and embedded versions of Java aren't free at all.
There is effectively only one Java implementation, the one controlled by Sun/Oracle. Sun killed most of the others early on with legal threats, and the few remaining ones seem to fail to meet the conditions of Sun's public patent grant.
Anybody who writes Java software is pretty much stuck with running it on Sun/Oracle's proprietary implementation or its nominally GPL derivative. You're joined at the hip with Oracle, in the bending over kind of sense.
It was Sun who never submitted Java to ISO or ANSI, it was Sun who created a dual-licensed Java, it was Sun who filed hundreds of patents on Java-related technologies, and it was Sun who created the limited patent grant under conditions that nobody could meet.
And it was predictable that Sun would eventually fail and get bought by someone who might start to enforce those patents; in fact, the reason Oracle was willing to outbid everybody else was probably because they realized that these patents hadn't been placed fully into the public domain.
I had been warning about this for years, but all the Java fanboys were arguing that Sun was the good guys, that they would never sue, and that Java was a free and open platform.
Do your homework people: what has happened was predictable, and the evil seeds were sown by Sun itself.
Of course, Google in using Java in the first place for android, is debatable, still
Google isn't really "using Java"; they are using the Java language, but almost none of the implementation or libraries.
Why did they choose the Java language? Because they needed a safe, statically typed, garbage collected language that people had experience with and that there were tools for. There is little else out there that fits the bill (C# wasn't an option at the time they started).
So much for all the people who said that Java was open, free, and not patent-encumbered. The Java patent grant set up conditions that you can essentially not meet unless you use Sun/Oracle's version. And the fact that Sun was going to be taken over was obvious for years. I had just hoped it was going to be IBM, who wouldn't have done this sort of thing.
Is math and is covered by mathematics degree programs are completely orthogonal properties.
Nevetheless, it is computer scientists, not mathematicians, that are gathering to to vet this paper: if you work on the theory of computation, you are a computer scientist, no matter what your training, background, or methods.
Correct: learning to program is a strict subset of computer science. (CPU and VLSI design may or may not have branched off from computer science now, that's debatable.)
There's a REASON that most of the algorithms we use every day were originally designed by people like Donald Knuth - who are professional MATHEMATICIANS.
Donald Knuth was trained as a mathematician before computer science existed, but today he is a computer scientists at the computer science department. Go check his home page.
In fact there is a methodology of programming that's gaining popularity for high-reliability tasks that does exactly that- writing programs by creating mathematical proofs first.
Be that as it may, it's computer scientists, not mathematicians, that are performing these proofs.
You're starting with the wrong premise; you're assuming that because people use mathematical reasoning and proofs, they are mathematicians. That was true 100 years ago, but not today. Today, there are several distinct mathematical fields. You don't go to the math department to study computer science, algorithms, or theory of computation. Hence, while computer science is a mathematical field, it's not the same as mathematics.
everything a computer does is just math and can be translated into any equivalent math and you'll get the same result
Being able to simulate and model something at a low level doesn't give you a high level understanding. A mathematician understands boolean logic, but that doesn't mean he has the knowledge and skills to design a CPU or program it. That knowledge and those skills aren't taught in mathematics, they are taught in computer science.
Ultimately it's easilly provable that every computer program is a simple mathematical function - so simple in fact that it is in fact a single number.
Every biological system is a physical system, but if you study physics, you'll know next to nothing about biology.
All computer science did was to create a (very shallow) layer of pretense through which ot access the maths.
There is no "pretense". Theoretical computer science is a highly mathematical discipline, but it deals with issues that are of no interest to most other mathematicians. They aren't part of the standard math curriculum either. However, the kinds of mathematics it deals with are of interest to people who build and program computers. That's why theoretical computer science became part of computer science and the computer science curriculum.
Lisp is DIRECTLY based on lambda-calculus - in fact the ONLY (minor) difference as small syntactical changes designed to make it easier to TYPE
Lisp has side effects, lexical scoping, a rich set of data types, and dynamic typing. It is nothing like lambda calculus.
Expect this to change in the next few years - multi-CPU machines are actually EASIER to program for in a functional language like lisp - which makes all those nasty multithreading issues just go away by putting you on the actual mathematics that happens.
Lisp is not a functional programming language, it's a multi-paradigm language that allows you to program in a functional style. Lisp programs are no more parallelizable than C or Perl programs, even when they are written in a functional style.
In order to be a functional programming language, a language needs to prohibit (or at least isolate) side effects. Haskell is about the closest to a pure functional programming language in common use that there is.
I think the fact that you think that Lisp has the benefits of lambda calculus illustrates again why mathematicians shouldn't be doing computer science...
But computer science is mathematics, or rather a subset thereof.
No, it's not. The relationship between computer science and mathematics is similar to the relationship between physics and mathematics: there is some overlap, lots of fruitful cooperation, and some fundamental differences in methodology and subject matter.
Even if computer science were a subset, it's only the computer scientists that are gathering to vet this paper. Saying "mathematicians are gathering to vet this paper" would be as misleading and irrelevant as saying "mammals are gathering to vet this paper".
It is certainly possible to patent a novel way of doing something that is not itself novel.
Yes, but Apple doesn't seem to have done that here.
I am not qualified to evaluate the novelty of Apple's claims.
Patents aren't supposed to be some mysterious, complex, legalistic scheme designed to give companies monopolies, they are supposed to communicate novel, non-obvious technical insights to engineers. If a working mobile software developer doesn't clearly see a significant novel, non-obvious technical insight in Apple's patent, then the patent should not be considered valid.
What is clear is that it would be a waste of time and money for Apple to try to patent ideas or user interfaces from apps, because any claim based on such a patent would immediately be thrown out of court on grounds of prior art
No patent gets "thrown out immediately" in court; you have expert witnesses, juries, and endless processes and procedures. A company that has a thousand bogus patents still has a strong patent portfolio because they can basically tie up any competitor in endless litigation. And that's roughly what Apple has.
But if they don't exist in many countries are they really that universal?
They are recognized by most nations.
What can you do about it beside bitch on the internet?
There is an international court of human rights, there are sanctions, and you can sue in the US, UK, and other nations for violations anywhere. Enforcement of universal rights is, of course, still difficult, but it's not impossible.
First of all, cryptography doesn't, in general, rely on an assumption that P!=NP.
P!=NP is related to public key cryptography, but not in a simple way. Public key cryptosystems relying on NP hard problems have failed because NP hardness isn't a guarantee that problem instances are hard to solve. On the other hand, current public key cryptosystems rely on factoring, which isn't even known to be NP complete. P=NP would result in polynomial algorithms for factoring, but not necessarily in a way that would break public key systems in any real way.
And that's another problem with P/NP considerations: (1) they are worst case complexity classes, and (2) the degree of P is not limited. As a result, problems in P may be impossibly hard to solve, while many instances of an NP hard problem may be easy to solve--too many to let you construct a good crypto system.
The GP wondered how long it would take for Apple to lock down OS X like they did with iPhone; he didn't claim that they already did. It's a valid concern given Apple's history. After all, in addition to what they have been doing with iPhone, they already restrict the OS to specific hardware, limit X11/desktop integration, and have deliberately made their SCSI drivers incompatible with non-Apple hardware.
This is statistics from a single university. Virginia was also one of the few places that used racks of XServe computers for supercomputing, and was showcased by Apple for that. Obviously, the university has some kind of special relationship with Apple. At the universities I have seen, Apple machines are a distant third to Windows and Linux, with Linux dominating in computer science. Most Mac usage seems to be laptops, not desktops.
There are some universal rights. Among them should be that no government should have the right to go on fishing expeditions through private communications. That isn't universally recognized, but hopefully we can get there.
Of course, the nations of the Middle East (Saudi Arabia, Egypt, Israel, Jordan, etc.) all commit far more serious human rights violations than merely digging through people's Blackberry messages, but still...
Schmidt had better kept his opinions to himself, given the kind of hot water that Google is in with privacy advocates. Instead of predicting the end of anonymity, he should apply the resources of his company to make sure anonymity remains possible, because his company is the most likely to suffer if, oh, Google Docs and Google Mail can't be used anonymously anymore. Unlike those services, OpenOffice on my desktop still is anonymous.
You have a blessedly naive idea of how companies like Apple work. People don't get "inspired" to write patents at companies because they like to, it's part of their job. They get rewarded if they write them, and they get punished if they don't. So, people figure out what they can get away with patenting that doesn't require too much of their time to do.
Without such strong management pressure, there is really no incentive to write patents: you don't see a financial reward from it, your company probably doesn't benefit from it much either, they just cause you problems when you switch jobs, and they take away weeks from productive work (or months if you're unlucky enough and your company gets involved in a lawsuit over that patent).
Just look at the filing date, December 18. This is the typical crappy end-of-year we-have-to-fill-our-quota patent. And while they didn't steal the idea from "Where To?", they did steal it from Tripit and similar services, many of which were around before that filing date.
In fact, the people who wrote that patent probably didn't even care that it's a bad patent; they filled their quota and made their Apple management happy, Jobs gets bragging rights over the size of his patent package, and if the patent doesn't stand up in court due to prior art, it's no skin off their backs.
Slashdot Mirror
Google have implemented 1 and 2 according to the specs.
No, Android does not implement the JVM. Android's Dalvik VM is a register based machine, just like the CLR. In part, that's probably because they wanted to minimize the risk of getting successfully sued by Sun. Let's hope they did their homework.
Not even Sun/Oracle is insane enough to propose that you should use the java standard edition on mobile phones.
No, instead they want you to license their embedded Java implementation, which isn't covered even by the hokey J2SE specs/licenses.
It's their problem, not Sun's.
It's also your and my problem if we rely on Java. That's why we shouldn't rely on Java.
Then why are there so many FOSS implementations?
Because Sun/Oracle hasn't bothered suing them out of existence yet. With its patent portfolio, Oracle has the power to kill pretty much any unlicensed implementation of Java.
Java is a spec and NOT an implementation. You are free to make your own implementation of the spec (IBM, Apple and many others do).
IBM and Apple have not "made their own implementations"; they have licensed Sun/Oracle's implementations and created derivatives.
You are not free to make your own implementation of the spec; you need to pass Sun's compatibility tests if you don't want to get hit by patent lawsuits because Sun hold essential patents for creating a conforming implementation.
For years now, there has been no implementation of Java conforming to the Java spec except for those derived from Sun's source code. That's not an accident: it's pretty much impossible to meet Sun's compatibility requirements without licensing their source code.
This would not be true if Apple cheated and did not implement some part of Java spec (which is the case with what Google did).
Google didn't "cheat", Google implemented their own platform and runtime; they just happened to use the Java language to do it. In principle, Sun/Oracle couldn't have done anything about that: Sun doesn't hold a patent on the Java language itself. But it appears as if the Android designers may not have been careful enough to avoid all of Sun's patents.
No, this isn't "like any other company" or like any other patent portfolio. Sun claimed that Java was an open platform when, in fact, it was highly patent encumbered, by design.
And open source developers had other choices. They chose Java because they erroneously believed that Sun's platform was open.
My comment above was a comment about Java, not Android.
What you quote about Android is correct in that it isn't an implementation of the Java platform. However, the patents that Oracle now owns do apply to Android if Android uses the techniques described in those patents (which I believe it does).
Of course it's intentional. It means that Sun effectively retained full control over all Java implementations, because the only implementations that could meet those criteria were the ones passing their compatibility tests and (in practice) only those based on their licensed source code. That makes Java a highly proprietary platform.
And where is your "run anywhere" going to go if Oracle decides not to renew Java licenses to Apple or IBM anymore? I don't think Apple or IBM can use the GPL version of Java on their platforms because they need to link non-GPL code into the implementation.
Java is, and always has been, a proprietary platform, in the same sense as Microsoft Windows has been. All Sun wanted to do was insert their proprietary layer in between the operating system and applications, meaning that you would be dependent on them instead of Microsoft. Looks like they may have the legal means to make it all stick.
and also because Java is licensed under the GPL.
"Java" isn't licensed under the GPL. A single Java implementation, derived from Sun's proprietary source code, is licensed under the GPL. Furthermore, the patent grant applies only if you meet specific compatibility conditions, which no implementation other than Sun's meets.
Google implemented the Java language, not its libraries, and did it by themselves. Android (and Dalvik) are licensed under a mix of Apache and GPL, but that doesn't matter; the license under which a third party implementation is released is not relevant for the patent grant.
Google rolled their own implementation and libraries for good reason: the full Java platform would have been far too obese for Android, and embedded versions of Java aren't free at all.
There is effectively only one Java implementation, the one controlled by Sun/Oracle. Sun killed most of the others early on with legal threats, and the few remaining ones seem to fail to meet the conditions of Sun's public patent grant.
Anybody who writes Java software is pretty much stuck with running it on Sun/Oracle's proprietary implementation or its nominally GPL derivative. You're joined at the hip with Oracle, in the bending over kind of sense.
It was Sun who never submitted Java to ISO or ANSI, it was Sun who created a dual-licensed Java, it was Sun who filed hundreds of patents on Java-related technologies, and it was Sun who created the limited patent grant under conditions that nobody could meet.
And it was predictable that Sun would eventually fail and get bought by someone who might start to enforce those patents; in fact, the reason Oracle was willing to outbid everybody else was probably because they realized that these patents hadn't been placed fully into the public domain.
I had been warning about this for years, but all the Java fanboys were arguing that Sun was the good guys, that they would never sue, and that Java was a free and open platform.
Do your homework people: what has happened was predictable, and the evil seeds were sown by Sun itself.
Of course, Google in using Java in the first place for android, is debatable, still
Google isn't really "using Java"; they are using the Java language, but almost none of the implementation or libraries.
Why did they choose the Java language? Because they needed a safe, statically typed, garbage collected language that people had experience with and that there were tools for. There is little else out there that fits the bill (C# wasn't an option at the time they started).
So much for all the people who said that Java was open, free, and not patent-encumbered. The Java patent grant set up conditions that you can essentially not meet unless you use Sun/Oracle's version. And the fact that Sun was going to be taken over was obvious for years. I had just hoped it was going to be IBM, who wouldn't have done this sort of thing.
Is math and is covered by mathematics degree programs are completely orthogonal properties.
Nevetheless, it is computer scientists, not mathematicians, that are gathering to to vet this paper: if you work on the theory of computation, you are a computer scientist, no matter what your training, background, or methods.
learning programming != computer science
Correct: learning to program is a strict subset of computer science. (CPU and VLSI design may or may not have branched off from computer science now, that's debatable.)
There's a REASON that most of the algorithms we use every day were originally designed by people like Donald Knuth - who are professional MATHEMATICIANS.
Donald Knuth was trained as a mathematician before computer science existed, but today he is a computer scientists at the computer science department. Go check his home page.
In fact there is a methodology of programming that's gaining popularity for high-reliability tasks that does exactly that- writing programs by creating mathematical proofs first.
Be that as it may, it's computer scientists, not mathematicians, that are performing these proofs.
You're starting with the wrong premise; you're assuming that because people use mathematical reasoning and proofs, they are mathematicians. That was true 100 years ago, but not today. Today, there are several distinct mathematical fields. You don't go to the math department to study computer science, algorithms, or theory of computation. Hence, while computer science is a mathematical field, it's not the same as mathematics.
everything a computer does is just math and can be translated into any equivalent math and you'll get the same result
Being able to simulate and model something at a low level doesn't give you a high level understanding. A mathematician understands boolean logic, but that doesn't mean he has the knowledge and skills to design a CPU or program it. That knowledge and those skills aren't taught in mathematics, they are taught in computer science.
Ultimately it's easilly provable that every computer program is a simple mathematical function - so simple in fact that it is in fact a single number.
Every biological system is a physical system, but if you study physics, you'll know next to nothing about biology.
All computer science did was to create a (very shallow) layer of pretense through which ot access the maths.
There is no "pretense". Theoretical computer science is a highly mathematical discipline, but it deals with issues that are of no interest to most other mathematicians. They aren't part of the standard math curriculum either. However, the kinds of mathematics it deals with are of interest to people who build and program computers. That's why theoretical computer science became part of computer science and the computer science curriculum.
Lisp is DIRECTLY based on lambda-calculus - in fact the ONLY (minor) difference as small syntactical changes designed to make it easier to TYPE
Lisp has side effects, lexical scoping, a rich set of data types, and dynamic typing. It is nothing like lambda calculus.
Expect this to change in the next few years - multi-CPU machines are actually EASIER to program for in a functional language like lisp - which makes all those nasty multithreading issues just go away by putting you on the actual mathematics that happens.
Lisp is not a functional programming language, it's a multi-paradigm language that allows you to program in a functional style. Lisp programs are no more parallelizable than C or Perl programs, even when they are written in a functional style.
In order to be a functional programming language, a language needs to prohibit (or at least isolate) side effects. Haskell is about the closest to a pure functional programming language in common use that there is.
I think the fact that you think that Lisp has the benefits of lambda calculus illustrates again why mathematicians shouldn't be doing computer science...
Producing oil is part of the normal function of human skin. If your skin doesn't do it, you're either a robot or very, very sick.
But computer science is mathematics, or rather a subset thereof.
No, it's not. The relationship between computer science and mathematics is similar to the relationship between physics and mathematics: there is some overlap, lots of fruitful cooperation, and some fundamental differences in methodology and subject matter.
Even if computer science were a subset, it's only the computer scientists that are gathering to vet this paper. Saying "mathematicians are gathering to vet this paper" would be as misleading and irrelevant as saying "mammals are gathering to vet this paper".
It is certainly possible to patent a novel way of doing something that is not itself novel.
Yes, but Apple doesn't seem to have done that here.
I am not qualified to evaluate the novelty of Apple's claims.
Patents aren't supposed to be some mysterious, complex, legalistic scheme designed to give companies monopolies, they are supposed to communicate novel, non-obvious technical insights to engineers. If a working mobile software developer doesn't clearly see a significant novel, non-obvious technical insight in Apple's patent, then the patent should not be considered valid.
What is clear is that it would be a waste of time and money for Apple to try to patent ideas or user interfaces from apps, because any claim based on such a patent would immediately be thrown out of court on grounds of prior art
No patent gets "thrown out immediately" in court; you have expert witnesses, juries, and endless processes and procedures. A company that has a thousand bogus patents still has a strong patent portfolio because they can basically tie up any competitor in endless litigation. And that's roughly what Apple has.
But if they don't exist in many countries are they really that universal?
They are recognized by most nations.
What can you do about it beside bitch on the internet?
There is an international court of human rights, there are sanctions, and you can sue in the US, UK, and other nations for violations anywhere. Enforcement of universal rights is, of course, still difficult, but it's not impossible.
P=NP is a question in computer science, not mathematics. The two fields split about 50 years ago.
No, that's wrong on many levels.
First of all, cryptography doesn't, in general, rely on an assumption that P!=NP.
P!=NP is related to public key cryptography, but not in a simple way. Public key cryptosystems relying on NP hard problems have failed because NP hardness isn't a guarantee that problem instances are hard to solve. On the other hand, current public key cryptosystems rely on factoring, which isn't even known to be NP complete. P=NP would result in polynomial algorithms for factoring, but not necessarily in a way that would break public key systems in any real way.
And that's another problem with P/NP considerations: (1) they are worst case complexity classes, and (2) the degree of P is not limited. As a result, problems in P may be impossibly hard to solve, while many instances of an NP hard problem may be easy to solve--too many to let you construct a good crypto system.
The GP wondered how long it would take for Apple to lock down OS X like they did with iPhone; he didn't claim that they already did. It's a valid concern given Apple's history. After all, in addition to what they have been doing with iPhone, they already restrict the OS to specific hardware, limit X11/desktop integration, and have deliberately made their SCSI drivers incompatible with non-Apple hardware.
This is statistics from a single university. Virginia was also one of the few places that used racks of XServe computers for supercomputing, and was showcased by Apple for that. Obviously, the university has some kind of special relationship with Apple. At the universities I have seen, Apple machines are a distant third to Windows and Linux, with Linux dominating in computer science. Most Mac usage seems to be laptops, not desktops.
There are some universal rights. Among them should be that no government should have the right to go on fishing expeditions through private communications. That isn't universally recognized, but hopefully we can get there.
Of course, the nations of the Middle East (Saudi Arabia, Egypt, Israel, Jordan, etc.) all commit far more serious human rights violations than merely digging through people's Blackberry messages, but still...
Schmidt had better kept his opinions to himself, given the kind of hot water that Google is in with privacy advocates. Instead of predicting the end of anonymity, he should apply the resources of his company to make sure anonymity remains possible, because his company is the most likely to suffer if, oh, Google Docs and Google Mail can't be used anonymously anymore. Unlike those services, OpenOffice on my desktop still is anonymous.
You have a blessedly naive idea of how companies like Apple work. People don't get "inspired" to write patents at companies because they like to, it's part of their job. They get rewarded if they write them, and they get punished if they don't. So, people figure out what they can get away with patenting that doesn't require too much of their time to do.
Without such strong management pressure, there is really no incentive to write patents: you don't see a financial reward from it, your company probably doesn't benefit from it much either, they just cause you problems when you switch jobs, and they take away weeks from productive work (or months if you're unlucky enough and your company gets involved in a lawsuit over that patent).
Just look at the filing date, December 18. This is the typical crappy end-of-year we-have-to-fill-our-quota patent. And while they didn't steal the idea from "Where To?", they did steal it from Tripit and similar services, many of which were around before that filing date.
In fact, the people who wrote that patent probably didn't even care that it's a bad patent; they filled their quota and made their Apple management happy, Jobs gets bragging rights over the size of his patent package, and if the patent doesn't stand up in court due to prior art, it's no skin off their backs.