"Then, too, if Windows Vista is any indication of what lies ahead, the company's software will continue to require ever more awe-inspiring hardware--a far cry from the light and nimble Web-based applications Mozilla engineers envision"
Despite the alleged lightness and nimbleness of web apps, they're still slower and more unreliable than native apps, when they work at all.
We're going around in circles. You're just making the same argument in a slightly different form. If financial considerations are equivalent or just as powerful as legal ones, than I suggest dropping the GPL in favor of a financial way of promoting F/OSS.
They could put a system in place that requires an ID and a signature before posting. This might destroy the value of YouTube but it's certainly possible.
"Second, like I said, it's viral in that if you use a commercially-licensed library in your own software (particularly one that charges a per-unit-shipped royalty), you have to release your software under a commercial license as well."
You don't have to use a commercial license for your own code and it's not even a financial problem except in those rather rare cases where you pay a per-unit royalty.
In any case, I don't agree that mere existance of incentives puts commerical projects in the viral category. The lack of choice is the key characteristic.
Nothing you quote in the MS license restricts you from releasing your own code under any license you choose, unless of course, the license you want to use requires that MS's source code is released as well. That would be giving away something you don't own.
"There is a symmetry: closed source begets closed source, and open source begets open source."
I guess that's true as long as you exclude the BSD license from the category of "open source". A curious classification particularly since there's been plenty of BSD code in GPL'd code.
In any case, the symmetry that you propose (regardless of it's logical flaws) has nothing to do with the "viral" issue. I can still release all of my original source code even if it uses closed source libraries, but I can't keep my original source code private if I link it to GPL'd code.
We can debate the value of each approach but it's silly to try to deny the fundamental nature of the GPL and how it differs from the closed source approach. After all, it's this behavior that some call viral that gives the GPL all it's power.
"Infection" simply means "you have to obey the licence terms".
No, not really. It's the nature of the license, not the mere fact that there is one. I don't see why it's so important to you to try to equate things that clearly aren't the same.
Viral has nothing to do with payment. If I write software for Linux I'm just as tied to it as you would be for a Windows app.
Viral refers to GPL'd code "infecting" other people's code when it is linked in. As others have correctly noted, it's a voluntary infection (since you don't have to use GPL'd code) but it is what it is.
"Commercial licenses are also viral, most of licenses do NOT allow you to redistribute/resell products using their tools/librairies, unless you pay an extravagant fee."
And this is viral how? Whether you believe the GPL is viral or not, the fundemental difference is that commerical licenses don't require you to distribute your indepedently written source code even if it's based on their libraries. As far as fees are concerned, many allow you to use their libraries simply because you paid for the tool with no per machine license.
I don't see what the big deal is. I put books in quotes because the original religious documents (assuming they were created when they were supposed to be) were written on scrolls, not bound in a book as we would think of it today. In fact if the name "Bible" really comes from the word book, the name is probably many centuries newer than the book itself.
"What was the reason for this not happening before again?"
A perceived lack of ROI, I would guess. Whether that perception was accurate will be determined once they've been offering the Linux PC's for a while and can weigh their profits against support costs.
The problem with the evolution vs. creationism debate is that the two don't form a dichotomy. If you could prove evolution was 100% wrong it wouldn't be any evidence that creationism was true.
Although science sometimes goes through periods of consensus on somewhat incorrect theories, there's a mechanism in science to correct and refine theories over time. Relativity, quantum mechanics, string theory, etc would not exist if science truly had the "one-right-way" viewpoint you attribute to it.
Religion on the other hand is based on faith and "books" that aren't going to change over time. There is no mechanism for self-correction because the truths believed in are supposed to be fixed for all time. Historically those who held different views were forced to start their own religion because they couldn't change the old one.
Some Christians can't believe in evolution, etc, because they fear it might mean there is no God. In those cases where science (using the same methods) produces theories that aren't God-threatening they don't mind. That's hardly surprising. Once the church thought that Galileo's theories were a threat to their faith, so they denounced them and him. Now that they don't think it's a threat anymore, his theories have become magically valid.
"It's also quite trendy and fashionable these days to attack the FSF and the GPL for "not being free" by some other definition that people keep changing."
No, it's the same, old, everyday definition of freedom that been around for hundreds of years.
"The ECMA documents state that the standards documents contain patented technology."
I never questioned that. The question is whether the patents that the standards refer to can be violated soley by implementing a version of.net It's quite possible that many open source projects violate these patents. Most companies try to make their patents as broad as possible to maximize their ability to collect license fees or to just lock out as many competitors as possible. The only way to be sure is to examine the specific patents.
Of course there may be issues such as prior art, but they'll apply equally to mono and other open (and closed) projects.
I don't know if there's a.net version of Ruby and of course I never claimed there was. Can we conclude that the JVM isn't capable of running other languages just because I can identify a language that won't compile to Java byte codes?
The fact is there are many languages that can cross-inherit in.net. This is a useful feature that some pro-Java people are in denial about.
So what are the patents in question that apply only to someone implementing.net and nothing else? If you are so certain that mono has a problem, you should be able to tell us the patent numbers.
As I said in another post, these patents are likely to be stated in the broadest language possible, not something like "patent on.NET implementation #638". Thus mono isn't likely to be in more danger than any other project.
Slashdot Mirror
"Then, too, if Windows Vista is any indication of what lies ahead, the company's software will continue to require ever more awe-inspiring hardware--a far cry from the light and nimble Web-based applications Mozilla engineers envision"
Despite the alleged lightness and nimbleness of web apps, they're still slower and more unreliable than native apps, when they work at all.
We're going around in circles. You're just making the same argument in a slightly different form. If financial considerations are equivalent or just as powerful as legal ones, than I suggest dropping the GPL in favor of a financial way of promoting F/OSS.
They could put a system in place that requires an ID and a signature before posting. This might destroy the value of YouTube but it's certainly possible.
"Second, like I said, it's viral in that if you use a commercially-licensed library in your own software (particularly one that charges a per-unit-shipped royalty), you have to release your software under a commercial license as well."
You don't have to use a commercial license for your own code and it's not even a financial problem except in those rather rare cases where you pay a per-unit royalty.
In any case, I don't agree that mere existance of incentives puts commerical projects in the viral category. The lack of choice is the key characteristic.
Nothing you quote in the MS license restricts you from releasing your own code under any license you choose, unless of course, the license you want to use requires that MS's source code is released as well. That would be giving away something you don't own.
That may explain how closed source could be expensive, it doesn't explain how it is viral.
"There is a symmetry: closed source begets closed source, and open source begets open source."
I guess that's true as long as you exclude the BSD license from the category of "open source". A curious classification particularly since there's been plenty of BSD code in GPL'd code.
In any case, the symmetry that you propose (regardless of it's logical flaws) has nothing to do with the "viral" issue. I can still release all of my original source code even if it uses closed source libraries, but I can't keep my original source code private if I link it to GPL'd code.
We can debate the value of each approach but it's silly to try to deny the fundamental nature of the GPL and how it differs from the closed source approach. After all, it's this behavior that some call viral that gives the GPL all it's power.
"Infection" simply means "you have to obey the licence terms".
No, not really. It's the nature of the license, not the mere fact that there is one. I don't see why it's so important to you to try to equate things that clearly aren't the same.
Viral has nothing to do with payment. If I write software for Linux I'm just as tied to it as you would be for a Windows app.
Viral refers to GPL'd code "infecting" other people's code when it is linked in. As others have correctly noted, it's a voluntary infection (since you don't have to use GPL'd code) but it is what it is.
"Commercial licenses are also viral, most of licenses do NOT allow you to redistribute/resell products using their tools/librairies, unless you pay an extravagant fee."
And this is viral how? Whether you believe the GPL is viral or not, the fundemental difference is that commerical licenses don't require you to distribute your indepedently written source code even if it's based on their libraries. As far as fees are concerned, many allow you to use their libraries simply because you paid for the tool with no per machine license.
I don't see what the big deal is. I put books in quotes because the original religious documents (assuming they were created when they were supposed to be) were written on scrolls, not bound in a book as we would think of it today. In fact if the name "Bible" really comes from the word book, the name is probably many centuries newer than the book itself.
"What was the reason for this not happening before again?"
A perceived lack of ROI, I would guess. Whether that perception was accurate will be determined once they've been offering the Linux PC's for a while and can weigh their profits against support costs.
The problem with the evolution vs. creationism debate is that the two don't form a dichotomy. If you could prove evolution was 100% wrong it wouldn't be any evidence that creationism was true.
Although science sometimes goes through periods of consensus on somewhat incorrect theories, there's a mechanism in science to correct and refine theories over time. Relativity, quantum mechanics, string theory, etc would not exist if science truly had the "one-right-way" viewpoint you attribute to it.
Religion on the other hand is based on faith and "books" that aren't going to change over time. There is no mechanism for self-correction because the truths believed in are supposed to be fixed for all time. Historically those who held different views were forced to start their own religion because they couldn't change the old one.
Some Christians can't believe in evolution, etc, because they fear it might mean there is no God. In those cases where science (using the same methods) produces theories that aren't God-threatening they don't mind. That's hardly surprising. Once the church thought that Galileo's theories were a threat to their faith, so they denounced them and him. Now that they don't think it's a threat anymore, his theories have become magically valid.
"It's also quite trendy and fashionable these days to attack the FSF and the GPL for "not being free" by some other definition that people keep changing."
No, it's the same, old, everyday definition of freedom that been around for hundreds of years.
"After the deal there is a good chance that Novell won't be so careful because they no longer have to fear legal action."
MS hasn't made any deal with Novell that shields it from a MS lawsuit.
"The ECMA documents state that the standards documents contain patented technology."
.net It's quite possible that many open source projects violate these patents. Most companies try to make their patents as broad as possible to maximize their ability to collect license fees or to just lock out as many competitors as possible. The only way to be sure is to examine the specific patents.
I never questioned that. The question is whether the patents that the standards refer to can be violated soley by implementing a version of
Of course there may be issues such as prior art, but they'll apply equally to mono and other open (and closed) projects.
Believe whatever you want. MS won't be able to sue anyone based on what ECMA documents say, it all depends on the wording of specific patents.
Sorry it's your claim that mono has a patent problem, you do the work.
I willing to consider your argument if you provide patent numbers that mono will violate and others won't.
I don't know if there's a .net version of Ruby and of course I never claimed there was. Can we conclude that the JVM isn't capable of running other languages just because I can identify a language that won't compile to Java byte codes?
.net. This is a useful feature that some pro-Java people are in denial about.
The fact is there are many languages that can cross-inherit in
So what are the patents in question that apply only to someone implementing .net and nothing else? If you are so certain that mono has a problem, you should be able to tell us the patent numbers.
"Did you miss the deal MS made with Novell to agree not to sue them due to patent concerns?"
I did miss that, but I did hear about a deal MS made with Novell to agree not to sue each others customers over patents.
As I said in another post, these patents are likely to be stated in the broadest language possible, not something like "patent on .NET implementation #638". Thus mono isn't likely to be in more danger than any other project.
So if your create a class in Ruby can it by subclassed in Java and vice versa? That's the level of language independence that .Net provides.