You are confusing ID chips with what "Digital Angel" claims to have developed. ID chips (I have tried them) have a range of a few feet, are powered by RF energy from the read-out wand, and can do virtually nothing other than transmit a number back to you. ADSX claims that they have something that can receive GPS, communicate bidirectionally with distant stations, is powered by muscle, and is the size of a dime. That's a completely different thing.
Collars for animal tracking are pretty big, not the size you would want to implant. And even they have fairly limited communications abilities. Wireless animal ID chips that are implanted, on the other hand, have very limited capabilities and a range of at most a few feet.
As for GPS, the antennas themselves are fairly substantial. Even the GPS watch is a pretty big device. It can do 600 readings (10h at 1 reading per minute) with a CR2 lithium battery. And how well that thing works compared to a "real" GPS receiver remains to be seen.
Making a dime-size device capable of bidirectional communications even just with a cell site, incorporating GPS, performing body function monitoring, and in addition being powered by muscular energy still looks to me like it belongs in the realm of science fiction. The fact that the individual bits and pieces seem plausible ("I have heard..." and "people are doing something like that for...") is just the mark of a good tall story.
A dime sized receiver and transmitter capable of communicating with satellites? And all of that while implated under the skin, inside a conductive medium? Powered by electricity generated from muscles?
If all that is possible, why are our GPS receivers still cell phone sized and operate for only 18h on a bunch of standard batteries? Why aren't there lots of simple implantable medical monitors that monitor on a much smaller scale? Why do this in humans first, when there are so many applications in animals and property tracking? Even though devices are less regulated than drugs, what about human testing?
Take a look at the stock chart on Yahoo! and check out the associated news and insider stock activity, salaries of the CEO, etc. The whole thing seems pretty iffy to me.
Nor did I insinuate that you made any claims about administrative or financial benefits. Your proposed solution still nicely illustrates some of the hidden costs of adopting NT, since a lot of people use similar solutions. From an engineering point of view, a $350 "solution" to a problem with a $150 operating system may not be a solution at all if those $350 bust the budget.
Ah, wait, if you have "high-end" SCSI drives or use some other NTFS features, or use some older drives, you need ERD from the same company. That's actually $349/user.
That's $150/user licensing fee, plus the training and maintenance required with bringing in yet another piece of third party software. I hope you take that into account in any "total cost of ownership" calculations.
In fact, your response is pretty characteristic of why NT is such a bad deal: you need zillions of commercial third party packages for all sorts of eventualities, and you get nickled-and-dimed to death, not to mention all the time that is wasted on this stuff.
Yes, that's the basic problem with Microsoft: they don't believe that experience and skill matter. Instead, they try to substitute a basic cookbook approach. That shows in their own approach to software development and in what they churn out.
Microsoft is doing to software what MacDonald's did to food. Both are financially extremely successful, and both are ultimately neither tasty nor good for you.
Linux is a great game platform, and there is demand for games on it. But it isn't surprising that games aren't selling like this:
Most Linux games come out long after the corresponding Windows game. Since many Linux users can dual boot, they'll already have gotten the Windows version by the time the Linux version comes out.
Linux games are often a pain to install. Q3 on RH6.1 took some effort, I haven't been able tog et it to work at all after upgrading to RH6.2.
The Linux games available at our local computer store are: Quake 1, Quake 2 (actually off the shelves already), Quake 3 (excellent), Railroad Tycoon (got OK reviews), Myth II (not my cup of tea, but got good reviews), and Heretic 2 (awful). That kind of selection isn't sufficient to keep a gamer happy. So, everybody has to be able to run Windows for games anyway, and then they might as well keep buying all their games for Windows.
Linux won't become a widely used, popular game platform until many of the major games are released for Windows and Linux within a short timespan of each other (what about releasing for Linux first, for a change? good way to do QA before the Windows masses and Windows technical problems bite). As long as vendors can break even on the games, I hope they will keep porting. It may not be a money maker now, but once the market catches up, it will be.
There is a lot of software out there for UNIX/Linux, much more than for Windows CE. And much of that software was designed to run on machines that were less powerful than the iPAQ.
X11 as well has a long history of use on embedded devices and low-end machines. The fact that it is so configurable and policy-free means that many applications can be moved from the desktop to a handheld simply by using a window manager adapted to the handheld.
And Linux is easy to program for. Windows CE is not even Win32 compatible, and it's a very stripped down environment.
The web design on http://www.surfnwash.net/ is broken: a big background image, a big logo, and some text, all arranged on separate layers with absolute pixel coordinates. It takes a long time to load and the stuff doesn't work in Netscape or Mozilla (you just get a vertical stack of the three layers). It doesn't work in text-based browsers either (all the buttons are images and if there were text, it would come out in the wrong order).
This kind of web design is happening more and more, even on some big e-commerce sites. It's bad not only for Netscape users but for just about anybody who wants to or has to use some other kind of display device or browser: handhelds, wireless, people with disabilities.
In this case, Adobe GoLive is at fault, and its own home page is broken as well. Throw it out or at least complain to Adobe.
NASA is "building the highway to the stars". They are developing and deploying new propulsion technologies and new computer technologies. The cheapest and quickest way of doing that is with unmanned probes. If we ever want manned space travel, that will require a lot of additional work, risk, and expense.
The shuttle seems like one of the worst places to put money. It was expensive and questionable technology even when it was originally developed. NASA (and other space agencies) have already found better ways of lifting both objects and people into orbit. Hopefully, we'll see that kind of new technology deployed soon. But then, we still have the choice of using the launch capacity for glorified space tourism (shooting senators into orbit) or actual scientific purposes (unmanned probes, space telescopes, etc.).
The last two missions were comparatively low cost. They failed because they tried to accomodate stingy tax payers and cut corners. Obviously, not a good decision by NASA. NASA apparently forgot the engineering principle "pick any two" in their "cheaper, faster, better" motto.
They know how to get probes to other planets pretty reliably. If they go back to the tried and true, but expensive, methods, their missions will likely work as well as they used to.
A method for combining on a computer display an additional set of information into a group of
multiple sets of information needed on a recurring basis, comprising the steps of:
It goes on to talk about "selection indicators" (which would include tabs, but also allow for other selection methods--buttons, checkboxes, etc.) and using them to select "different information" to be displayed.
I see nothing about "customization" or "palettes" in those claims. They seem to apply very well to any use of tabbed widgets.
Of course, Adobe's legal strategy would be to go first to what is obviously most similar to their current product. That way, they can point to a concrete instance and claim harm from the "infringement". If that goes through, they can go after other "infringers" afterwards much more easily if they want to.
If Adobe has to use such marginal patents in tactical legal maneuvers to bolster its business, it looks to me like they must anticipate financial trouble. Stock analysts, take notice.
Beyond that, their pronouncements on intellectual property seem particularly hypocritical given that their core technology, PostScript, was developed by the Adobe founders while at Xerox PARC, and I doubt that Xerox got a lot of money out of Adobe for that.
Why would you guess? It takes only a few seconds to look up the claims on the IBM patent server. And the claims are broad enough to cover pretty arbitrary uses of tabbed widgets.
Will this stand up? Who knows. But patent applicants have no interest in narrowing down their claims--if they claim too much, the courts will simply narrow it down for them to as much as they can get away with.
Much as I dislike the RIAA, I have trouble seeing why copyright should revert to artists in the recording industry. If you are an engineer and develop software, hardware, or an invention as part of your job, ownership of the ideas doesn't magically revert to you after a few dozen years. And I don't see why it should. Why would it be any different for artists?
I think we shouldn't carve out special interest niches in copyright law for writers, artists, or other folks, we should fix copyright law as a whole and bring copyright terms back to something reasonable.
That's a common idea but pretty naive. The concept of "own time" generally doesn't exist in professional corporate employment (at least in the US). Anything you do, at any time, that is even vaguely related to your job function, automatically belongs to your employer. The same is likely true for a student, if not for any other reason, because the notion of "on your own time" is rather hard to define for students.
Your view of the world is that the practical products your university makes pay for the impractical pursuits you engage in at the university.
Well, that simply can't work in the long run. From the point of view of paying for the practical products, you, your research, and your education are pure overhead. Any entity competing with Columbia that doesn't have to pay that kind overhead will simply produce the same product (practical inventions) at a lower cost.
Universities produce public goods--education and basic scientific research--and those must be paid for either by the government or by donations. If, on the other hand, universities start relying primarily on student tuition and corporate funding, they'll turn into a different kind of institution: trade schools, corporate training schools, and development labs. If Columbia wants to turn into those, fine, but why bother? There are already enough of them around.
Even if we stay within the framework of free market economic theory, the theory recognizes the existence of something called "public goods". The free market will not efficiently create those. And basic research (discoveries that can't be made proprietary or patented) is one of the textbook examples of public goods. Left exclusively to free market mechanisms, a lot of important research just isn't going to get done because it is in no individual's economic interest to pay for it.
So, your analysis is fatally flawed even within a dogmatic free market view.
Of course, there are plenty of other reasons not to subject specific areas of economic activity to free market mechanisms, because we may have other considerations and goals that are not linked to the efficient use of resources.
NeXTStep and Objective-C ought to have caught on in the 1980's: they were a good compromise between practicality and powerful language features. It would have been great if that had been widely used. Looking back, it's not too hard to see where NeXT and Jobs went wrong.
In 2000, Objective-C is still nice in many ways but it's dated. Java receives all the "hype" because it is at the sweet spot of backwards compatibility and new features right now. Most importantly, Java is runtime safe and it has well-defined reflection, something Objective-C always lacked, and something that is essential for a modern component market. I think Apple is making the right decision in going strongly after Java in MacOS X.
Static member variables or static enquiry methods, are entirely adequate for that. The JavaBeans framework uses them, for example. If the language has reflection, they become even more powerful.
C# attributes seem like a lot of messy mechanism for very little functionality. They should redirect that effort to add a decent reflection facility.
I think it's a mistake to assume that the preferred ML implementation of genericity is better (or worse) than C++ style genericity. Both have their areas of application.
C++ style genericity is unmatched in its expressiveness and efficiency for scientific computations. If you are going to have value classes (also called "expanded classes"), you need C++ style genericity with it to use them for anything useful; without that, they become nearly meaningless. An ML-style implementation doesn't even come close in providing the necessary functionality.
On the other hand, C++ style genericity is cumbersome and inefficient for general purpose "application programming". For that, the ML style implementation (and GenericJava) genericity is much better.
I hope the GenericJava folks are not working under the assumption that the ML strategy is always preferable, because if they do, what they produce will likely be useless for high performance applications. Compiler-created specializations of generic classes, including type variable specific overloading, and value classes are the backbone of high level, high performance computing.
But the law doesn't talk about "intellectual PROPERTY", it talks about "patents", "copyrights", and "trademarks". The term "intellectual property" is a convenient (and misleading) shorthand to refer to that diverse collection of laws that grant certain people special rights for a limited amount of time.
Objective-C wasn't really "revolutionary". When it came out, Smalltalk had already been around for many years, there were Lisp machines, and lots of excellent languages and environments. All the important GUI and graphics concepts of NeXTStep also had existed in other environment before.
Objective-C was a pragmatic attempt to bring at least some of that functionality to a world already dominated by a systems programming language called "C". The biggest problem with Objective-C was (and continues to be) that it inherited the unsafe nature of C.
I think it would have been good for the industry for Objective-C and NeXT to catch on. It would have put the industry on a different trajectory. If Objective-C had been used widely, by now, the language would probably have evolved to be significantly safer, with C-like functionality restricted to specific, unsafe modules where needed.
But the NeXT system did evolve and has found widespread acceptance, not at NeXT/Apple but at Sun. The closest predecessor to Java is not C++ but it's Objective-C: Java has much of the dynamic binding, reflection, and library from Objective-C and the NeXT. That's not really surprising either, since many of the people working on Java for the last few years came from the Smalltalk/Self and from the Objective-C communities.
Slashdot Mirror
You are confusing ID chips with what "Digital Angel" claims to have developed. ID chips (I have tried them) have a range of a few feet, are powered by RF energy from the read-out wand, and can do virtually nothing other than transmit a number back to you. ADSX claims that they have something that can receive GPS, communicate bidirectionally with distant stations, is powered by muscle, and is the size of a dime. That's a completely different thing.
As for GPS, the antennas themselves are fairly substantial. Even the GPS watch is a pretty big device. It can do 600 readings (10h at 1 reading per minute) with a CR2 lithium battery. And how well that thing works compared to a "real" GPS receiver remains to be seen.
Making a dime-size device capable of bidirectional communications even just with a cell site, incorporating GPS, performing body function monitoring, and in addition being powered by muscular energy still looks to me like it belongs in the realm of science fiction. The fact that the individual bits and pieces seem plausible ("I have heard..." and "people are doing something like that for...") is just the mark of a good tall story.
If all that is possible, why are our GPS receivers still cell phone sized and operate for only 18h on a bunch of standard batteries? Why aren't there lots of simple implantable medical monitors that monitor on a much smaller scale? Why do this in humans first, when there are so many applications in animals and property tracking? Even though devices are less regulated than drugs, what about human testing?
Take a look at the stock chart on Yahoo! and check out the associated news and insider stock activity, salaries of the CEO, etc. The whole thing seems pretty iffy to me.
Nor did I insinuate that you made any claims about administrative or financial benefits. Your proposed solution still nicely illustrates some of the hidden costs of adopting NT, since a lot of people use similar solutions. From an engineering point of view, a $350 "solution" to a problem with a $150 operating system may not be a solution at all if those $350 bust the budget.
Ah, wait, if you have "high-end" SCSI drives or use some other NTFS features, or use some older drives, you need ERD from the same company. That's actually $349/user.
In fact, your response is pretty characteristic of why NT is such a bad deal: you need zillions of commercial third party packages for all sorts of eventualities, and you get nickled-and-dimed to death, not to mention all the time that is wasted on this stuff.
Microsoft is doing to software what MacDonald's did to food. Both are financially extremely successful, and both are ultimately neither tasty nor good for you.
- Most Linux games come out long after the corresponding Windows game. Since many Linux users can dual boot, they'll already have gotten the Windows version by the time the Linux version comes out.
- Linux games are often a pain to install. Q3 on RH6.1 took some effort, I haven't been able tog et it to work at all after upgrading to RH6.2.
- The Linux games available at our local computer store are: Quake 1, Quake 2 (actually off the shelves already), Quake 3 (excellent), Railroad Tycoon (got OK reviews), Myth II (not my cup of tea, but got good reviews), and Heretic 2 (awful). That kind of selection isn't sufficient to keep a gamer happy. So, everybody has to be able to run Windows for games anyway, and then they might as well keep buying all their games for Windows.
Linux won't become a widely used, popular game platform until many of the major games are released for Windows and Linux within a short timespan of each other (what about releasing for Linux first, for a change? good way to do QA before the Windows masses and Windows technical problems bite). As long as vendors can break even on the games, I hope they will keep porting. It may not be a money maker now, but once the market catches up, it will be.X11 as well has a long history of use on embedded devices and low-end machines. The fact that it is so configurable and policy-free means that many applications can be moved from the desktop to a handheld simply by using a window manager adapted to the handheld.
And Linux is easy to program for. Windows CE is not even Win32 compatible, and it's a very stripped down environment.
Furthermore, these kinds of rules do often apply to students, both in the US and abroad. If you ignore them, at least know what you are doing.
This kind of web design is happening more and more, even on some big e-commerce sites. It's bad not only for Netscape users but for just about anybody who wants to or has to use some other kind of display device or browser: handhelds, wireless, people with disabilities.
In this case, Adobe GoLive is at fault, and its own home page is broken as well. Throw it out or at least complain to Adobe.
The shuttle seems like one of the worst places to put money. It was expensive and questionable technology even when it was originally developed. NASA (and other space agencies) have already found better ways of lifting both objects and people into orbit. Hopefully, we'll see that kind of new technology deployed soon. But then, we still have the choice of using the launch capacity for glorified space tourism (shooting senators into orbit) or actual scientific purposes (unmanned probes, space telescopes, etc.).
They know how to get probes to other planets pretty reliably. If they go back to the tried and true, but expensive, methods, their missions will likely work as well as they used to.
It goes on to talk about "selection indicators" (which would include tabs, but also allow for other selection methods--buttons, checkboxes, etc.) and using them to select "different information" to be displayed. I see nothing about "customization" or "palettes" in those claims. They seem to apply very well to any use of tabbed widgets.
Of course, Adobe's legal strategy would be to go first to what is obviously most similar to their current product. That way, they can point to a concrete instance and claim harm from the "infringement". If that goes through, they can go after other "infringers" afterwards much more easily if they want to.
Beyond that, their pronouncements on intellectual property seem particularly hypocritical given that their core technology, PostScript, was developed by the Adobe founders while at Xerox PARC, and I doubt that Xerox got a lot of money out of Adobe for that.
Will this stand up? Who knows. But patent applicants have no interest in narrowing down their claims--if they claim too much, the courts will simply narrow it down for them to as much as they can get away with.
I think we shouldn't carve out special interest niches in copyright law for writers, artists, or other folks, we should fix copyright law as a whole and bring copyright terms back to something reasonable.
That's a common idea but pretty naive. The concept of "own time" generally doesn't exist in professional corporate employment (at least in the US). Anything you do, at any time, that is even vaguely related to your job function, automatically belongs to your employer. The same is likely true for a student, if not for any other reason, because the notion of "on your own time" is rather hard to define for students.
Well, that simply can't work in the long run. From the point of view of paying for the practical products, you, your research, and your education are pure overhead. Any entity competing with Columbia that doesn't have to pay that kind overhead will simply produce the same product (practical inventions) at a lower cost.
Universities produce public goods--education and basic scientific research--and those must be paid for either by the government or by donations. If, on the other hand, universities start relying primarily on student tuition and corporate funding, they'll turn into a different kind of institution: trade schools, corporate training schools, and development labs. If Columbia wants to turn into those, fine, but why bother? There are already enough of them around.
Of course, there are plenty of other reasons not to subject specific areas of economic activity to free market mechanisms, because we may have other considerations and goals that are not linked to the efficient use of resources.
In 2000, Objective-C is still nice in many ways but it's dated. Java receives all the "hype" because it is at the sweet spot of backwards compatibility and new features right now. Most importantly, Java is runtime safe and it has well-defined reflection, something Objective-C always lacked, and something that is essential for a modern component market. I think Apple is making the right decision in going strongly after Java in MacOS X.
C# attributes seem like a lot of messy mechanism for very little functionality. They should redirect that effort to add a decent reflection facility.
C++ style genericity is unmatched in its expressiveness and efficiency for scientific computations. If you are going to have value classes (also called "expanded classes"), you need C++ style genericity with it to use them for anything useful; without that, they become nearly meaningless. An ML-style implementation doesn't even come close in providing the necessary functionality.
On the other hand, C++ style genericity is cumbersome and inefficient for general purpose "application programming". For that, the ML style implementation (and GenericJava) genericity is much better.
I hope the GenericJava folks are not working under the assumption that the ML strategy is always preferable, because if they do, what they produce will likely be useless for high performance applications. Compiler-created specializations of generic classes, including type variable specific overloading, and value classes are the backbone of high level, high performance computing.
But the law doesn't talk about "intellectual PROPERTY", it talks about "patents", "copyrights", and "trademarks". The term "intellectual property" is a convenient (and misleading) shorthand to refer to that diverse collection of laws that grant certain people special rights for a limited amount of time.
Objective-C was a pragmatic attempt to bring at least some of that functionality to a world already dominated by a systems programming language called "C". The biggest problem with Objective-C was (and continues to be) that it inherited the unsafe nature of C.
I think it would have been good for the industry for Objective-C and NeXT to catch on. It would have put the industry on a different trajectory. If Objective-C had been used widely, by now, the language would probably have evolved to be significantly safer, with C-like functionality restricted to specific, unsafe modules where needed.
But the NeXT system did evolve and has found widespread acceptance, not at NeXT/Apple but at Sun. The closest predecessor to Java is not C++ but it's Objective-C: Java has much of the dynamic binding, reflection, and library from Objective-C and the NeXT. That's not really surprising either, since many of the people working on Java for the last few years came from the Smalltalk/Self and from the Objective-C communities.