"I often hear this claim with respect to the GPL - but has there ever been a case where a judge disagreed with the distribution terms, and declared it public domain?"
That's not the only possible scenario. Although unlikely, a court could strike down parts of the GPL while maintaining that the remaining parts are still in force.
I don't think allocation is the issue we were discussing, but deallocation. My point is that manual deallocation doesn't require any "discovery" time to figure out which object should be deallocated. It's the delay in deallocation that has the potential of keeping excess memory in play.
A well-written C++ program is going to free memory much faster than a GC can. The value of GC is that you don't have to worry about forgetting to free memory, it will happen - eventually.
"Most GUI apps perform tasks that can be done better with a programming or markup language. Compare Word to LaTeX, for instance."
This is an entirely different subject. You could use a GUI to create a LaTeX document or a CLI editor to create a Word document (and while we're at it, what is it about UNIX culture that creates a desire to elevate file-formats to "languages").
"And frankly I would be browsing with text if more sites were properly designed. "
I'd say that the www adds no value if you're going to use text only. Telnet provides all you would need.
"If GUIs were actually a superior way of interfacing with a computer, then we'd be programming with them."
Programming applications is an entirely different activity than using applications. If we are going to pretend that all activities on a computer are equivalent, we could also say that if a CLI were a superior way of interfacing with a computer, then we'd be browsing with them. After all, if humans are "built for language" then a numeric representation of graphics should be more intuitive than a graphical image.
"Humans are built for language, and interacting with the computer with a language is orders of magnitude more simple, direct, and intuitive than some physical metaphor"
I think if you do a little research you'd find that humans have been manipulating the physical world for thousands of years before languages came along. Written languages came along much, much later. In addition CLI "languages" have little in common with any spoken language.
Of course, the research on learning and retention that GUI's are based on are well established and predate the CLI vs. GUI argument. Believe it or not, GUI's weren't invented just to piss off Unix folks.
Re:got Mono - stay away or risk infection w/MS ger
on
GNOME 2.20 Released
·
· Score: 1
Yes, the patent issue with respect to mono is nonsense. MS (or any other legally-wise company) is going to make their patents as broad as possible and not tie them to a specific implementation. Thus non-mono code is just as likely to trip over MS patents.
An absurd alternate-universe example: if email were patented by MS and implemented only in.NET, than using that feature in your mono program might get you in trouble. On the other hand, if you decided to create a non-mono version of email for Linux, you'd still be in trouble because the patent is based on the email "invention" not on the specific implementation.
I can't say I buy into this argument, but in any case, the GPL doesn't fully support it. Under the GPL, non-distributed code can be changed without those changes remaining in the "gene pool". Even in the case where the code is distributed, it's quite possible that the original project leaders will never discover the changes or never decide to incorporate them if they do encounter them. So the whole "gene pool" scenario doesn't really apply to the GPL.
"Sorry, but winning a stupid argument about OS naming isn't an important enough reason to finish a non-trivial software project (especially when you can go play with research concepts on it instead)."
Well, that's fine if one were to decide to lose the argument and stop bitching, but that doesn't seem to be the case. Besides, if one avoids doing a project on the basis that it's non-trival, there must not be much pride or ambition involved.
"The OpenBSD people are married to Unix and C, in much the same way as Bjarne believes the answer is to fix C++, and not to use Java and C#."
I think Bjarne's real position is that no one tool is right for everything. His general comment about computer languages is that as they mature and become fully functional, the initial simplicity that everyone got excited about evaporates.
You don't have to look any further than Java generics to see the truth of that (my observation, not Bjarne's).
"So, the question is not about what I think or what Eben thinks. It is what a judge would decide if a case came down to it."
Yes, that's the point I was trying to make. Some seem to have the idea that the opinion of the writer of an license carries some legal weight: as if RMS could be called as an expert witness on the GPL to tell the court what it really means. The whole point of writing something down is to make the document embody the meaning rather than individual opinions.
Understood, but the whole article is meaningless if you interpret it to only apply to "pleasure" use. Apple would have to make significant inroads in the busines market to overtake Windows.
"Why is your legal opinion on the issue even slightly relevant?
Eben Moglen is a lawyer who has been asked for legal advice on this issue by programmers to whom it directly matters and has provided an informed legal opinion."
Since neither Moglen nor the GP is a judge deciding the legalities of this issue, neither of their opinions is relevant from a legal point of view. Moglen as a lawyer could potentially be better informed than the GP, but that doesn't mean his opinion caries any more legal weight.
Re:%75 as effective as a prescription 3% the price
on
Science vs. Homeopathy
·
· Score: 1
"But in fact, you haven't fixed anything psychologically, because you continue to have a mindset such that you respond inflexibly when you have to face certain realities."
I'm not sure that depression could be defined as "inflexibly when you have to face certain realities", but in any case, if therapy is effective it has to be making changes to the brain which is what antidepressants do as well. The difference is that the drugs are more consistently effective at doing it than therapy is.
Re:%75 as effective as a prescription 3% the price
on
Science vs. Homeopathy
·
· Score: 2
"Also, we can pretty much write off Prozac because it has become the Ritalin of middle-age. By that I mean that a wide array of causes, behavioral, social, or chemical, are causing a problem, and instead of resolving it (through behavioral therapy or psychological analysis) the doc is just writing for the same treatment. Bobby is loud, give him Adderall. Bobby is sad, give him Prozac. Some people really need the chemically altering action of Prozac to be happy- some people just want to buy a month's worth of 10mg Problem Solver from CVS... i digress."
How exactly does behavioral therapy or physiological analysis "resolve" a problem? If your wife left you, will therapy bring her back? No, only the way you feel can be addressed. The fundamental problem will never be resolved. I would be thoughtful before taking a drug like Prozac, but I'm not swayed by the unscientific protestant-ethic-based theory that solving a problem should be hard or time-consuming.
"I think the GP wasn't aware that the GGP was being facetious."
Actually, I was aware but I was suggesting that his mocking wasn't entirely appropriate. TCP/IP is a key technology behind the Internet, but if it cost a minimum of $5000 - $10000 to buy a computer capable of browsing the web (because there was no standard platform to amortize costs) than it wouldn't matter. The Internet would still be a government and university experiment.
You did misunderstand. The sentence about TCP/IP in the first paragraph didn't have anything to do with the second paragraph. Perhaps my post could have been organized better. On the other hand, who would refer to TCP/IP as a "standard platform"?
I don't believe that only MS, IBM and Intel could have provided a popular standard platform that made the web practical from a business perspective, but somebody had to. Perhaps it could have been Apple instead, but it certainly wouldn't have been UNIX or it's clones.
Well, you should avoid libraries that are OS dependent if possible. Thus you can use the best combination of tools considering both their cost and their quality. Of course GPL'd code may not be in the running if you wish to keep your own code private.
"How are you going to perform when your boss is popping in your office every half hour asking if you fixed the [insert some error] that client B is experiencing preventing them from doing their job which is costing them 10k an hour."
It would be simpler to just ask: "How are you going to perform if you work for a badly managed company"? Answer: "don't".
Of course, employers do the same kind of dumb things. They have a job posting with a long list of obscure technologies the applicant is supposed to have experience with, but in the summary they say the job requires 0-2 years experience. This is how companies can claim that they can't find qualified people: by creating requirements that are impossible to fill.
Slashdot Mirror
"I often hear this claim with respect to the GPL - but has there ever been a case where a judge disagreed with the distribution terms, and declared it public domain?"
That's not the only possible scenario. Although unlikely, a court could strike down parts of the GPL while maintaining that the remaining parts are still in force.
"Keeping data around is again no problem, because the memory limit for useful data can be manually specified"
Sorry, I don't get your point. Keeping data around when it's no longer needed seems to be exactly the problem we're discussing.
I don't think allocation is the issue we were discussing, but deallocation. My point is that manual deallocation doesn't require any "discovery" time to figure out which object should be deallocated. It's the delay in deallocation that has the potential of keeping excess memory in play.
A well-written C++ program is going to free memory much faster than a GC can. The value of GC is that you don't have to worry about forgetting to free memory, it will happen - eventually.
than at least one person attending MIT.
"Most GUI apps perform tasks that can be done better with a programming or markup language. Compare Word to LaTeX, for instance."
This is an entirely different subject. You could use a GUI to create a LaTeX document or a CLI editor to create a Word document (and while we're at it, what is it about UNIX culture that creates a desire to elevate file-formats to "languages").
"And frankly I would be browsing with text if more sites were properly designed. "
I'd say that the www adds no value if you're going to use text only. Telnet provides all you would need.
"Ever try to walk someone through something when you have to tell them where exactly to click on a screen you can't see? "
"Every try to enter a CLI command on somebody else's non-networked computer when you can't reach their keyboard?
"If GUIs were actually a superior way of interfacing with a computer, then we'd be programming with them."
Programming applications is an entirely different activity than using applications. If we are going to pretend that all activities on a computer are equivalent, we could also say that if a CLI were a superior way of interfacing with a computer, then we'd be browsing with them. After all, if humans are "built for language" then a numeric representation of graphics should be more intuitive than a graphical image.
"Humans are built for language, and interacting with the computer with a language is orders of magnitude more simple, direct, and intuitive than some physical metaphor"
I think if you do a little research you'd find that humans have been manipulating the physical world for thousands of years before languages came along. Written languages came along much, much later. In addition CLI "languages" have little in common with any spoken language.
Of course, the research on learning and retention that GUI's are based on are well established and predate the CLI vs. GUI argument. Believe it or not, GUI's weren't invented just to piss off Unix folks.
Yes, the patent issue with respect to mono is nonsense. MS (or any other legally-wise company) is going to make their patents as broad as possible and not tie them to a specific implementation. Thus non-mono code is just as likely to trip over MS patents.
.NET, than using that feature in your mono program might get you in trouble. On the other hand, if you decided to create a non-mono version of email for Linux, you'd still be in trouble because the patent is based on the email "invention" not on the specific implementation.
An absurd alternate-universe example: if email were patented by MS and implemented only in
I can't say I buy into this argument, but in any case, the GPL doesn't fully support it. Under the GPL, non-distributed code can be changed without those changes remaining in the "gene pool". Even in the case where the code is distributed, it's quite possible that the original project leaders will never discover the changes or never decide to incorporate them if they do encounter them. So the whole "gene pool" scenario doesn't really apply to the GPL.
"Sorry, but winning a stupid argument about OS naming isn't an important enough reason to finish a non-trivial software project (especially when you can go play with research concepts on it instead)."
Well, that's fine if one were to decide to lose the argument and stop bitching, but that doesn't seem to be the case. Besides, if one avoids doing a project on the basis that it's non-trival, there must not be much pride or ambition involved.
"The OpenBSD people are married to Unix and C, in much the same way as Bjarne believes the answer is to fix C++, and not to use Java and C#."
I think Bjarne's real position is that no one tool is right for everything. His general comment about computer languages is that as they mature and become fully functional, the initial simplicity that everyone got excited about evaporates.
You don't have to look any further than Java generics to see the truth of that (my observation, not Bjarne's).
"So, the question is not about what I think or what Eben thinks. It is what a judge would decide if a case came down to it."
Yes, that's the point I was trying to make. Some seem to have the idea that the opinion of the writer of an license carries some legal weight: as if RMS could be called as an expert witness on the GPL to tell the court what it really means. The whole point of writing something down is to make the document embody the meaning rather than individual opinions.
Understood, but the whole article is meaningless if you interpret it to only apply to "pleasure" use. Apple would have to make significant inroads in the busines market to overtake Windows.
And if there's one thing that can really make your business, it's putting a video on YouTube in 5 min instead of 10.
"Why is your legal opinion on the issue even slightly relevant?
Eben Moglen is a lawyer who has been asked for legal advice on this issue by programmers to whom it directly matters and has provided an informed legal opinion."
Since neither Moglen nor the GP is a judge deciding the legalities of this issue, neither of their opinions is relevant from a legal point of view. Moglen as a lawyer could potentially be better informed than the GP, but that doesn't mean his opinion caries any more legal weight.
"But in fact, you haven't fixed anything psychologically, because you continue to have a mindset such that you respond inflexibly when you have to face certain realities."
I'm not sure that depression could be defined as "inflexibly when you have to face certain realities", but in any case, if therapy is effective it has to be making changes to the brain which is what antidepressants do as well. The difference is that the drugs are more consistently effective at doing it than therapy is.
"Also, we can pretty much write off Prozac because it has become the Ritalin of middle-age. By that I mean that a wide array of causes, behavioral, social, or chemical, are causing a problem, and instead of resolving it (through behavioral therapy or psychological analysis) the doc is just writing for the same treatment. Bobby is loud, give him Adderall. Bobby is sad, give him Prozac. Some people really need the chemically altering action of Prozac to be happy- some people just want to buy a month's worth of 10mg Problem Solver from CVS... i digress."
How exactly does behavioral therapy or physiological analysis "resolve" a problem? If your wife left you, will therapy bring her back? No, only the way you feel can be addressed. The fundamental problem will never be resolved. I would be thoughtful before taking a drug like Prozac, but I'm not swayed by the unscientific protestant-ethic-based theory that solving a problem should be hard or time-consuming.
"I think the GP wasn't aware that the GGP was being facetious."
Actually, I was aware but I was suggesting that his mocking wasn't entirely appropriate. TCP/IP is a key technology behind the Internet, but if it cost a minimum of $5000 - $10000 to buy a computer capable of browsing the web (because there was no standard platform to amortize costs) than it wouldn't matter. The Internet would still be a government and university experiment.
You did misunderstand. The sentence about TCP/IP in the first paragraph didn't have anything to do with the second paragraph. Perhaps my post could have been organized better. On the other hand, who would refer to TCP/IP as a "standard platform"?
The only key technology you mention is TCP/IP.
I don't believe that only MS, IBM and Intel could have provided a popular standard platform that made the web practical from a business perspective, but somebody had to. Perhaps it could have been Apple instead, but it certainly wouldn't have been UNIX or it's clones.
Well, you should avoid libraries that are OS dependent if possible. Thus you can use the best combination of tools considering both their cost and their quality. Of course GPL'd code may not be in the running if you wish to keep your own code private.
"How are you going to perform when your boss is popping in your office every half hour asking if you fixed the [insert some error] that client B is experiencing preventing them from doing their job which is costing them 10k an hour."
It would be simpler to just ask: "How are you going to perform if you work for a badly managed company"? Answer: "don't".
Of course, employers do the same kind of dumb things. They have a job posting with a long list of obscure technologies the applicant is supposed to have experience with, but in the summary they say the job requires 0-2 years experience. This is how companies can claim that they can't find qualified people: by creating requirements that are impossible to fill.