Oh, I know you can add your own. I was just making an observation that the G4s ship with so much good stuff, and then they throw in this substandard (by today's measure) graphics card.:)
Apple is friendly to the LinuxPPC developers, because Linux on Mac doesn't represent a threat to them.
To Apple, BeOS on Mac *DOES* represent a threat, so they have become stingy. They're well within their rights to do so, so I'm not faulting them for that... I just wish it wasn't so.
Otherwise, I might have been a lot more willing to take the money I sank into a new system last year and buy myself a Macintosh (or, the money I just sank into a Dell and buy a PowerBook).
Be's problem isn't with the processors themselves; it's with the motherboards/chipsets that the processors run on, and those motherboards/chipsets are Apple's responsibility.
Apple refuses to let Be use/buy the specifications for their motherboards/chipsets, so as a result Be is effectively shut-out from the Macintosh G3/G4 computer system. Be *could* get BeOS up and working on an alternate motherboard system (if a third-party started making CHRP clones, for instance), but again you run into the problems of market share (if you thought the market share for alternative operating systems on native Apple hardware was small, just wait until you see the market share for alternative OSes on alternative hardware).
Of course, the fact that Apple desperately wanted to woo Steve Jobs back into CEO position and buying NeXT would've been a great way to do that has absolutely nothing to do with the purchase.
PowerPC is open to those whom Apple is willing to help.
Apple refuses to allow Be to get the specifications necessary for BeOS to run on the G3/G4 processors. So, officially, BeOS is shut out of the later PowerPC processors. Unofficially, Be could reverse-engineer the data necessary to get BeOS up and running, or borrow it from the various LinuxPPC projects.
The problem with that idea is that BeOS is a commercial company with finite resources. It would take them a lot of effort to bring the Mac version of BeOS up to date on the G3/G4 processors and would not necessarily provide them with much benefit - currently, the market for alternative operating systems on the Macintosh is miniscule. Due to their limited human resources (i.e., engineers), their effort is better off spent on improving the operating system on Intel rather than retard overall improvement to bring it up to date for the Mac.
Furthermore, this isn't a one-time thing: it would be an on-going concern. They would have to continually split their developers, and whenever Apple decided on a whim to swap things around in their motherboards, make changes in their ROM, etc., Be would be right back where they started - wasting engineering effort to get things fixed. Linux, to a large degree, doesn't have this problem - much of the LinuxPPC work is done by volunteers who aren't wasting their employer's money. Furthermore, despite all of the noise to the contrary Apple doesn't really see Linux as a competitor to their main operating system, so Apple has been willing to sort of slip the LinuxPPC developers information now and then. Apple does, however, clearly see BeOS as an OS competitor.
On top of that, Be would also have to worry about possible lawsuits. If things like CSS can be considered trade secrets, God only knows what can be considered as viable material for grounds in a lawsuit. Code in the ROM, interaction with various processor components - you name it, I'm sure Apple has half-a-dozen things they could sue over. Whereas "part-time" Linux hackers don't represent a worthwhile target (since Apple doesn't see them as competition), you bet your ass that Apple would sue BeOS if BeOS ever actually become worthwhile competition.
In other words, Apple doesn't want them on their platform and Be, Inc. has a lot of reasons not to challenge that stance. They continually ask Apple, but Apple continually refuses. There's not much more they can do with taking very great risks to their business.
Did it ever occur to you that it might be an oversight?
Perhaps they did read the license. Perhaps, in releasing the code for the other numerous pieces of GPL software that Be uses, they simply overlooked this one (or it fell through the cracks).
Not every slight to the GPL is intentional, you know.
Actually, sir AC, I have grown up and I do "see the light." I simply fail to see what your point is.
FreeBSD can't really die, for the same reason that Linux can't really die. The nature of their respective open source licenses allows for the concept of forking. If enough people who work on FreeBSD don't like the new direction of the merged FreeBSD/BSDI parent company, they can simply take the latest copy of the pure FreeBSD source code and do their own thing.
Yes, the BSD license allows BSDI to take the FreeBSD code, merge into into their operating system, and keep the resulting changes proprietary. But that's the "price" you pay for using the BSD license. You allow people to keep their changes proprietary, and you accept that. But, as before, that doesn't mean that FreeBSD suddenly becomes proprietary as well. At worst, the evolution of FreeBSD may be stunted, but it won't die off if something drastic were to happen to the FreeBSD/BSDi merger.
As for your fears of OS cannibalism, mixing the two operating systems explicitly involves tearing down the two original systems and building a single, integrated system. After the two have been integrated, you won't have the same FreeBSD you have right now. That's the price of progress. The chance you have to take is that the FreeBSD/BSDi merger is going to integrate the best aspects of both operating systems to produce a superior OS and that the parent companies feel it's in their best interest to publish the code to their changes.
For one thing, it's not really switching "from a complete closed source model, to a Free/Open Source software model."
They're basically taking FreeBSD and using that as a base for building a better FreeBSD. Over the next few years, they're going to selectively take BSDI code and merge it in, and release the resulting merged code as BSD licensed. That's not really "open sourcing" BSDI, and it doesn't count as such.
On the other hand, they *ARE* pursuing a more aggressive open source model. Rather than open sourcing BSDI, and adding another fragment to the BSD world, they're going to merge the two operating systems to build a much better operating system than either alone. While you can't say that it's "really" open-sourcing BSDI, in many ways it's a step better.
Kudos to BSDI for doing this - they make a better product in the end AND they do it in a friendly, intelligent manner.
The problem is, even if you're right in saying that the gun was originally invented for killing, it has evolved beyond that point. There are many, many other uses for guns nowadays beyond sheer killing, and even much of the killing that *is* done with guns is discriminatory.
So what you're doing is flogging a dead horse. Realistically, I can justify owning a gun under the terms of personal protection, for the exact same reason police officers are given weapons: because there are criminals out there who do not understand anything but a loaded weapon. Criminals own weapons, and it has been demonstrated repeatedly in American society that virtually all criminals don't obey the laws regarding weapon ownership. Gun control does control guns, but it basically controls guns only for those people who abide by the laws - the same people who are LEAST likely to be the aggressors in a gun-related incident.
As a result, while you've made life difficult for a small number of criminals who purchase guns, by and large you have not made an impact in the overall ownership of firearms by the criminal populace. What you HAVE done is denied the ordinary, intelligent, well-trained citizen of their best defense against the well-armed criminal population. "Rounding up guns," as you said in another post, doesn't do anything but deny the general public of a legitimate, efficient method of self-defense.
Of course, kids like Klebold and Harris, or that sixth-grader in Michigan, don't have such qualms. They'll obtain firearms illegally because of adults who don't give a damn about your "silly little gun laws", or they'll find alternate ways of doing the harm they want to (Klebold and Harris built bombs from instructions off the Internet). The type of strict, harsh gun control you propose is a mental placebo - it makes you feel good, it makes you think that you're doing something to solve the problem, but when it comes right down to it you've solved nothing, because the criminals will continue to buy, sell, and use firearms against the now defenseless populace.
It's legislative and political masturbation. Instead of solving the actual problem, you're simply throwing a tarp over it, wiping your hands clean, and saying to yourself, "Now the problem no longer exists."
It can't be tested directly. In other words, you can't go out tomorrow and build yourself a wormhole and see if it's stable. What you *can* do is determine what this theory relies on - what types of matter does it require, the properties of that matter, the implications it has on other theories, and so on. In other words, few theories exist by themselves - they're the end results of "chains" of theories.
As time goes by, we find experimental evidence for many things. Sometimes it's macroscopic evidence (although not directly photographed, there's enough other types of evidence that black holes are pretty much considered physical proven objects), and other times it's microscopic evidence (such as the recent evidence that neutrinos may have mass after all).
By using these bits of evidence as building blocks, we can test the chains of a particular theory. By disproving - or failing to disprove - the theories a new theory rests on, then you can go a long way towards testing these ideas.
Furthermore, the key phrase you need to be aware of is "present technology." What may be beyond our current technology may not be beyond our future technology - should we wait until technology catches up to speculate about the universe? In many ways, it's the speculation and striving to determine the inner workings of the universe that dictate our technology and the direction it takes. In many instances, the theory has to come before the physical implementation - I sincerely doubt anyone could've built a nuclear reactor before understanding something about the way nuclear reactions occur.
Based on what I said, you reply changes nothing. A gun was designed for efficient killing, a pipe was designed for effcient transfer of water.
Ah, but you're still wrong.
The design of a gun achieves one function: to accelerate a lump of metal to a high speed in a very small period of time.
The design of a pipe achieves one function: allowing any object smaller than the diameter of the pipe to go in one end and come out at one or more other ends.
Neither one is inherently designed for killing. Period.
That's what AOL testified that they did - when somebody presented them with a complaint, they acted upon it.
The court decision seems to say that that isn't good enough - you have to be proactive and shut the stuff down *before* it starts. And, frankly, that's unfeasible at best and utterly impossible at worst.
The purpose of a tool is not inherent in its design (although a particular design may lend itself more towards a subset of purposes) - the purpose of a tool is determined by the user. If, in the absence of a hammer, I decide to use a large granite rock to bang in a nail, for the moment the purpose of that rock is as a make-shift hammer. If I need a battering ram, and the only thing I can find is a large tree that has fallen over, then the purpose of that fallen tree is as a battering ram. The purpose of an object shifts from use to use, and as a result cannot be said to have any one particular use.
In the United States, we arm our police officers with guns. If guns are such a travesty, why do we issue our police guns? Can't we simply hand them batons or tasers and go our separate ways? No, because the criminals have guns as well. So the guns we give our policemen become a defense. They become a deterrent to some types of crimes, and they become a defense against others.
Further, in all the rush for gun control, taking guns off the street, and banning guns altogether, I have yet to hear a single person say that once guns are abolished in American society that we should disarm our police officers as well. Why? Guns, for police officers, provide that deterrent/defense.
Where is the outcry? After all, if guns are designed to kill, shouldn't the police be going on massive shooting rampages? They carry guns everday - shouldn't we be hearing daily about a policeman simply snapping and gunning down a half-dozen people in a McDonalds? No, and the reason why is that we consider policemen to be responsible users of weapons. They have taken the training, they practice safe handling procedures, don't keep their guns near children, and so on.
Responsible gun owners do all of the above. They know how to fire a handgun because they have taken the training. They have an inherent notion of what constitutes safe handling procedures, and the responsible owners take great pains to secure their guns from their children - and to teach their children that guns are not a plaything. As long as I can remember, my father had guns in the house. I know that he had several different handguns and several different varieties of rifles. Yet he was a responsible gun owner, taught me right from wrong, and taught me that guns weren't play toys - and, as a result, I never even considered picking up a gun, no matter how bad my life was or how much somebody else irritated me.
So what we have here is a conflicting nature of guns: on the one hand, we sanction our police officers to carry guns because they have a positive purpose - deterrent against crime, and the ability of an officer to defend himself against a violent criminal. On the other hand, we have your implied argument, that the only things guns are good for are killing.
I take it, then, that you believe the police officers have the right to kill whenever and whoever they want, and that you don't have any problems with that.
"Better than A0 is the ultimate in security: the Heisenburg Security Rating. Here, for example, is a system rated as Heisenburg Secure. This locked box may or may not even contain a computer - and until you open it, it actually contains both! Of course, the network connections and power cables inside the box connect to the non-existent portion of the Heisenburg Secure system..."
Actually, there's an even more secure rating than A1.
A0, as defined by the military, is an unplugged, completely disassembled computer system where all volatile magnetic memory has been exposed to extremely powerful electromagnets for at least 72 hours. Each piece is then separately taken Cape Canaveral via several different types of transportation (horse, plane, submarine, postal carrier), launch into orbit (using Space Shuttle flights randomly chosen from a calendar) whereupon they are loaded into small, disposable rockets and fired towards the sun along with all documentation.
>If Java is so great at creating the programs that fill these niches, why is no vertical market app vendor shipping a Java product to do so?
Asked and answered. These niches that you refer to are inherently specific. While you can lump them up and say, for instance, "servlets for accessing a database," the particulars vary. Company XYZ may require an *entirely* different servlet than Company ABC, because of the configuration of their database, how they intend to access and display the retrieved data, and any number of other variables. Company XYZ could package that up and ship it, but it would be doubtful that anymore than a handful of other companies would find the software useful without serious modification.
The next step above that would be to sell a generic servlet. But again, you run into the configuration issue. How much do you have to architect the servlet before it becomes applicable to more than a small handful of situations? How bloated does it have to become to encompass the vast variety of configurations out there? Many corporations would find it easier to simply write their own from scratch, ensuring that the servlet would be completely designed for optimal performance for a particular configuration.
As a result, much of the Java work that is done is done for company-specific purposes, and is not sold or otherwise distributed because it would be of extremely limited value to others without the source code to modify. It would be like the United States trying to sell the programming instructions inside a Patriot missile. Unless you *have* a Patriot missile (raise your hands, please) the code isn't going to be of direct value to you, and if you happen to have another type of missile, the Patriot code isn't going to work without serious modifications.
If you want to see what exactly is being shipped, I suggest you look at Java Solutions Marketplace, where you can browse through the Java-related products and services to see whether anything qualifies for your stringent criteria.
>The only possible exception to the above are problems so crazy and stupid (and small) that no sane company would make a product to fix it.
You seem to want Java products that you can walk into a store and buy. It doesn't work like that. Should I dismiss Perl because I can't walk into CompUSA and buy a pure Perl software package? Perl has it's place as the glue that holds together the web. Java also has it's place in the software world - it's just not the visible, retail shelves that you are so desperate for.
>Now we're back to non-real-world uses.
I'd be interested in seeing how you define non-real-world uses. If it optimally solves a problem for a company, as far as I'm concerned that IS a real-world use (and is a much better real-world-use than some generic packaged product that has to be shoehorned into being merely an adequate solution).
Shipping products don't necessarily equate with wide usage. From the people I've talked to in the industry (and they represent a fair cross-section), Java is used overwhelmingly for internal use. Those coding efforts will never see the light of day publicly because they represent specific solutions to a specific internal problem. Many organizations use Java as sort of a glue-all; where Perl is considered the duct tape of the Internet, Java is more like a set of schematics and a trained team of engineers - Perl gets the job done, but Java can get the job done *right.*
What you're trying to do is equate Java's acceptance with the level of shipping products based on the language. Where your metric falls down - and will fall down with any language or utility - is if that language or utility isn't used in products that you can buy off the shelf. Java isn't used in many shipping products because it isn't suited very well towards client applications. Where it IS suited very well is on the server - a place of use that you miss completely by asking only for shipping products.
I didn't say it made sense.:) I simply said that's what Ticketmaster was complaining about.
I agree, Ticketmaster does make money on the sales. But when have you ever known a demonstrably greedy corporation like Ticketmaster to not pursue their attempts to squeeze the last dime out of the revenue stream through any means necessary, up to and including suing the living blazes out of anyone that gets in their way?
Slashdot Mirror
They can't. Sales, marketing, and legal are all one department. :)
What version of glibc did you use?
:)
Any particular specific comments as to why glibc is such a bear, and what you did about them?
(For someone considering rolling-his-own distro for laptops, I'm curious.
Frankly, after using both BeOS and OpenStep, I would've bought BeOS.
But hey, that's why I'm not a CEO and am instead relegated to the dreary realm of programmer...
Oh, I know you can add your own. I was just making an observation that the G4s ship with so much good stuff, and then they throw in this substandard (by today's measure) graphics card. :)
Point in case.
Apple is friendly to the LinuxPPC developers, because Linux on Mac doesn't represent a threat to them.
To Apple, BeOS on Mac *DOES* represent a threat, so they have become stingy. They're well within their rights to do so, so I'm not faulting them for that... I just wish it wasn't so.
Otherwise, I might have been a lot more willing to take the money I sank into a new system last year and buy myself a Macintosh (or, the money I just sank into a Dell and buy a PowerBook).
Pardon me for my oversimplification.
Be's problem isn't with the processors themselves; it's with the motherboards/chipsets that the processors run on, and those motherboards/chipsets are Apple's responsibility.
Apple refuses to let Be use/buy the specifications for their motherboards/chipsets, so as a result Be is effectively shut-out from the Macintosh G3/G4 computer system. Be *could* get BeOS up and working on an alternate motherboard system (if a third-party started making CHRP clones, for instance), but again you run into the problems of market share (if you thought the market share for alternative operating systems on native Apple hardware was small, just wait until you see the market share for alternative OSes on alternative hardware).
Then, of course, there's the issue of buying a decent monitor. :)
A good 19" monitor will cost you $500 to $600. That certainly doesn't bring it into $2500 range, but it certainly closes the gap somewhat.
On the other hand, those G4 systems offer fairly competitive hardware to IBM-compatible systems (aside from the crappy video card).
Of course, the fact that Apple desperately wanted to woo Steve Jobs back into CEO position and buying NeXT would've been a great way to do that has absolutely nothing to do with the purchase.
PowerPC is open to those whom Apple is willing to help.
Apple refuses to allow Be to get the specifications necessary for BeOS to run on the G3/G4 processors. So, officially, BeOS is shut out of the later PowerPC processors. Unofficially, Be could reverse-engineer the data necessary to get BeOS up and running, or borrow it from the various LinuxPPC projects.
The problem with that idea is that BeOS is a commercial company with finite resources. It would take them a lot of effort to bring the Mac version of BeOS up to date on the G3/G4 processors and would not necessarily provide them with much benefit - currently, the market for alternative operating systems on the Macintosh is miniscule. Due to their limited human resources (i.e., engineers), their effort is better off spent on improving the operating system on Intel rather than retard overall improvement to bring it up to date for the Mac.
Furthermore, this isn't a one-time thing: it would be an on-going concern. They would have to continually split their developers, and whenever Apple decided on a whim to swap things around in their motherboards, make changes in their ROM, etc., Be would be right back where they started - wasting engineering effort to get things fixed. Linux, to a large degree, doesn't have this problem - much of the LinuxPPC work is done by volunteers who aren't wasting their employer's money. Furthermore, despite all of the noise to the contrary Apple doesn't really see Linux as a competitor to their main operating system, so Apple has been willing to sort of slip the LinuxPPC developers information now and then. Apple does, however, clearly see BeOS as an OS competitor.
On top of that, Be would also have to worry about possible lawsuits. If things like CSS can be considered trade secrets, God only knows what can be considered as viable material for grounds in a lawsuit. Code in the ROM, interaction with various processor components - you name it, I'm sure Apple has half-a-dozen things they could sue over. Whereas "part-time" Linux hackers don't represent a worthwhile target (since Apple doesn't see them as competition), you bet your ass that Apple would sue BeOS if BeOS ever actually become worthwhile competition.
In other words, Apple doesn't want them on their platform and Be, Inc. has a lot of reasons not to challenge that stance. They continually ask Apple, but Apple continually refuses. There's not much more they can do with taking very great risks to their business.
Did it ever occur to you that it might be an oversight?
Perhaps they did read the license. Perhaps, in releasing the code for the other numerous pieces of GPL software that Be uses, they simply overlooked this one (or it fell through the cracks).
Not every slight to the GPL is intentional, you know.
Wow. Criticizing Bruce Perens about failing to advocate the GPL properly.
FascDot, you have *balls*...
"Grow up and see the light."
Actually, sir AC, I have grown up and I do "see the light." I simply fail to see what your point is.
FreeBSD can't really die, for the same reason that Linux can't really die. The nature of their respective open source licenses allows for the concept of forking. If enough people who work on FreeBSD don't like the new direction of the merged FreeBSD/BSDI parent company, they can simply take the latest copy of the pure FreeBSD source code and do their own thing.
Yes, the BSD license allows BSDI to take the FreeBSD code, merge into into their operating system, and keep the resulting changes proprietary. But that's the "price" you pay for using the BSD license. You allow people to keep their changes proprietary, and you accept that. But, as before, that doesn't mean that FreeBSD suddenly becomes proprietary as well. At worst, the evolution of FreeBSD may be stunted, but it won't die off if something drastic were to happen to the FreeBSD/BSDi merger.
As for your fears of OS cannibalism, mixing the two operating systems explicitly involves tearing down the two original systems and building a single, integrated system. After the two have been integrated, you won't have the same FreeBSD you have right now. That's the price of progress. The chance you have to take is that the FreeBSD/BSDi merger is going to integrate the best aspects of both operating systems to produce a superior OS and that the parent companies feel it's in their best interest to publish the code to their changes.
For one thing, it's not really switching "from a complete closed source model, to a Free/Open Source software model."
They're basically taking FreeBSD and using that as a base for building a better FreeBSD. Over the next few years, they're going to selectively take BSDI code and merge it in, and release the resulting merged code as BSD licensed. That's not really "open sourcing" BSDI, and it doesn't count as such.
On the other hand, they *ARE* pursuing a more aggressive open source model. Rather than open sourcing BSDI, and adding another fragment to the BSD world, they're going to merge the two operating systems to build a much better operating system than either alone. While you can't say that it's "really" open-sourcing BSDI, in many ways it's a step better.
Kudos to BSDI for doing this - they make a better product in the end AND they do it in a friendly, intelligent manner.
Misinformed? No. Just not nearly as closed-minded as you are.
The problem is, even if you're right in saying that the gun was originally invented for killing, it has evolved beyond that point. There are many, many other uses for guns nowadays beyond sheer killing, and even much of the killing that *is* done with guns is discriminatory.
So what you're doing is flogging a dead horse. Realistically, I can justify owning a gun under the terms of personal protection, for the exact same reason police officers are given weapons: because there are criminals out there who do not understand anything but a loaded weapon. Criminals own weapons, and it has been demonstrated repeatedly in American society that virtually all criminals don't obey the laws regarding weapon ownership. Gun control does control guns, but it basically controls guns only for those people who abide by the laws - the same people who are LEAST likely to be the aggressors in a gun-related incident.
As a result, while you've made life difficult for a small number of criminals who purchase guns, by and large you have not made an impact in the overall ownership of firearms by the criminal populace. What you HAVE done is denied the ordinary, intelligent, well-trained citizen of their best defense against the well-armed criminal population. "Rounding up guns," as you said in another post, doesn't do anything but deny the general public of a legitimate, efficient method of self-defense.
Of course, kids like Klebold and Harris, or that sixth-grader in Michigan, don't have such qualms. They'll obtain firearms illegally because of adults who don't give a damn about your "silly little gun laws", or they'll find alternate ways of doing the harm they want to (Klebold and Harris built bombs from instructions off the Internet). The type of strict, harsh gun control you propose is a mental placebo - it makes you feel good, it makes you think that you're doing something to solve the problem, but when it comes right down to it you've solved nothing, because the criminals will continue to buy, sell, and use firearms against the now defenseless populace.
It's legislative and political masturbation. Instead of solving the actual problem, you're simply throwing a tarp over it, wiping your hands clean, and saying to yourself, "Now the problem no longer exists."
It *can* be tested... sort of.
It can't be tested directly. In other words, you can't go out tomorrow and build yourself a wormhole and see if it's stable. What you *can* do is determine what this theory relies on - what types of matter does it require, the properties of that matter, the implications it has on other theories, and so on. In other words, few theories exist by themselves - they're the end results of "chains" of theories.
As time goes by, we find experimental evidence for many things. Sometimes it's macroscopic evidence (although not directly photographed, there's enough other types of evidence that black holes are pretty much considered physical proven objects), and other times it's microscopic evidence (such as the recent evidence that neutrinos may have mass after all).
By using these bits of evidence as building blocks, we can test the chains of a particular theory. By disproving - or failing to disprove - the theories a new theory rests on, then you can go a long way towards testing these ideas.
Furthermore, the key phrase you need to be aware of is "present technology." What may be beyond our current technology may not be beyond our future technology - should we wait until technology catches up to speculate about the universe? In many ways, it's the speculation and striving to determine the inner workings of the universe that dictate our technology and the direction it takes. In many instances, the theory has to come before the physical implementation - I sincerely doubt anyone could've built a nuclear reactor before understanding something about the way nuclear reactions occur.
Ah, yes. The armchair, untrained theorist who is smarter than every theoretical physicist on the rest of the Earth, including Stephen Hawking.
If you're so smart, why are you wasting your time on Slashdot?
Based on what I said, you reply changes nothing. A gun was designed for efficient killing, a pipe was designed for effcient transfer of water.
Ah, but you're still wrong.
The design of a gun achieves one function: to accelerate a lump of metal to a high speed in a very small period of time.
The design of a pipe achieves one function: allowing any object smaller than the diameter of the pipe to go in one end and come out at one or more other ends.
Neither one is inherently designed for killing. Period.
That's what AOL testified that they did - when somebody presented them with a complaint, they acted upon it.
The court decision seems to say that that isn't good enough - you have to be proactive and shut the stuff down *before* it starts. And, frankly, that's unfeasible at best and utterly impossible at worst.
Your notion is flawed.
The purpose of a tool is not inherent in its design (although a particular design may lend itself more towards a subset of purposes) - the purpose of a tool is determined by the user. If, in the absence of a hammer, I decide to use a large granite rock to bang in a nail, for the moment the purpose of that rock is as a make-shift hammer. If I need a battering ram, and the only thing I can find is a large tree that has fallen over, then the purpose of that fallen tree is as a battering ram. The purpose of an object shifts from use to use, and as a result cannot be said to have any one particular use.
In the United States, we arm our police officers with guns. If guns are such a travesty, why do we issue our police guns? Can't we simply hand them batons or tasers and go our separate ways? No, because the criminals have guns as well. So the guns we give our policemen become a defense. They become a deterrent to some types of crimes, and they become a defense against others.
Further, in all the rush for gun control, taking guns off the street, and banning guns altogether, I have yet to hear a single person say that once guns are abolished in American society that we should disarm our police officers as well. Why? Guns, for police officers, provide that deterrent/defense.
Where is the outcry? After all, if guns are designed to kill, shouldn't the police be going on massive shooting rampages? They carry guns everday - shouldn't we be hearing daily about a policeman simply snapping and gunning down a half-dozen people in a McDonalds? No, and the reason why is that we consider policemen to be responsible users of weapons. They have taken the training, they practice safe handling procedures, don't keep their guns near children, and so on.
Responsible gun owners do all of the above. They know how to fire a handgun because they have taken the training. They have an inherent notion of what constitutes safe handling procedures, and the responsible owners take great pains to secure their guns from their children - and to teach their children that guns are not a plaything. As long as I can remember, my father had guns in the house. I know that he had several different handguns and several different varieties of rifles. Yet he was a responsible gun owner, taught me right from wrong, and taught me that guns weren't play toys - and, as a result, I never even considered picking up a gun, no matter how bad my life was or how much somebody else irritated me.
So what we have here is a conflicting nature of guns: on the one hand, we sanction our police officers to carry guns because they have a positive purpose - deterrent against crime, and the ability of an officer to defend himself against a violent criminal. On the other hand, we have your implied argument, that the only things guns are good for are killing.
I take it, then, that you believe the police officers have the right to kill whenever and whoever they want, and that you don't have any problems with that.
*chuckle*
"Better than A0 is the ultimate in security: the Heisenburg Security Rating. Here, for example, is a system rated as Heisenburg Secure. This locked box may or may not even contain a computer - and until you open it, it actually contains both! Of course, the network connections and power cables inside the box connect to the non-existent portion of the Heisenburg Secure system..."
Actually, there's an even more secure rating than A1.
A0, as defined by the military, is an unplugged, completely disassembled computer system where all volatile magnetic memory has been exposed to extremely powerful electromagnets for at least 72 hours. Each piece is then separately taken Cape Canaveral via several different types of transportation (horse, plane, submarine, postal carrier), launch into orbit (using Space Shuttle flights randomly chosen from a calendar) whereupon they are loaded into small, disposable rockets and fired towards the sun along with all documentation.
You and I seem to be talking past each other.
>If Java is so great at creating the programs that fill these niches, why is no vertical market app vendor shipping a Java product to do so?
Asked and answered. These niches that you refer to are inherently specific. While you can lump them up and say, for instance, "servlets for accessing a database," the particulars vary. Company XYZ may require an *entirely* different servlet than Company ABC, because of the configuration of their database, how they intend to access and display the retrieved data, and any number of other variables. Company XYZ could package that up and ship it, but it would be doubtful that anymore than a handful of other companies would find the software useful without serious modification.
The next step above that would be to sell a generic servlet. But again, you run into the configuration issue. How much do you have to architect the servlet before it becomes applicable to more than a small handful of situations? How bloated does it have to become to encompass the vast variety of configurations out there? Many corporations would find it easier to simply write their own from scratch, ensuring that the servlet would be completely designed for optimal performance for a particular configuration.
As a result, much of the Java work that is done is done for company-specific purposes, and is not sold or otherwise distributed because it would be of extremely limited value to others without the source code to modify. It would be like the United States trying to sell the programming instructions inside a Patriot missile. Unless you *have* a Patriot missile (raise your hands, please) the code isn't going to be of direct value to you, and if you happen to have another type of missile, the Patriot code isn't going to work without serious modifications.
If you want to see what exactly is being shipped, I suggest you look at Java Solutions Marketplace, where you can browse through the Java-related products and services to see whether anything qualifies for your stringent criteria.
>The only possible exception to the above are problems so crazy and stupid (and small) that no sane company would make a product to fix it.
You seem to want Java products that you can walk into a store and buy. It doesn't work like that. Should I dismiss Perl because I can't walk into CompUSA and buy a pure Perl software package? Perl has it's place as the glue that holds together the web. Java also has it's place in the software world - it's just not the visible, retail shelves that you are so desperate for.
>Now we're back to non-real-world uses.
I'd be interested in seeing how you define non-real-world uses. If it optimally solves a problem for a company, as far as I'm concerned that IS a real-world use (and is a much better real-world-use than some generic packaged product that has to be shoehorned into being merely an adequate solution).
The problem is that your metric is faulty.
Shipping products don't necessarily equate with wide usage. From the people I've talked to in the industry (and they represent a fair cross-section), Java is used overwhelmingly for internal use. Those coding efforts will never see the light of day publicly because they represent specific solutions to a specific internal problem. Many organizations use Java as sort of a glue-all; where Perl is considered the duct tape of the Internet, Java is more like a set of schematics and a trained team of engineers - Perl gets the job done, but Java can get the job done *right.*
What you're trying to do is equate Java's acceptance with the level of shipping products based on the language. Where your metric falls down - and will fall down with any language or utility - is if that language or utility isn't used in products that you can buy off the shelf. Java isn't used in many shipping products because it isn't suited very well towards client applications. Where it IS suited very well is on the server - a place of use that you miss completely by asking only for shipping products.
I didn't say it made sense. :) I simply said that's what Ticketmaster was complaining about.
I agree, Ticketmaster does make money on the sales. But when have you ever known a demonstrably greedy corporation like Ticketmaster to not pursue their attempts to squeeze the last dime out of the revenue stream through any means necessary, up to and including suing the living blazes out of anyone that gets in their way?