No, absolutely wrong. Contract law is a seperate category and is not related to copyright law. A copyright license is different legal entity than a contract, and violating one is not a breach of contract - it is a totally seperate action, falling under different legislation, with different issues and different penalties.
Second, since you no longer have permission to use the copy of the software in your possession, you are in violation of federal copyright law
In the portion that I quoted, he vaguely referenced "federal law" for violating a license.
If you'd use that magic scrollbar thingy and look down a couple paragraphs you'd see where he mentions securities law. Maybe you should read the *whole* article before posting next time. Of course, even if you don't know how to work a scrollbar, theres this thing called "context" that might lead you to believe he's talking about Sarbanes-Oxley.
Next, I discussed how the Sarbanes-Oxley provision might apply, and why I didn't think that it did.
No, you made up some random case that wouldn't be covered by SOX and knocked down the strawman iwth great strength and power, totally ignoring the actual point made by the article.
It's not spammers, it's compromised machines used as (spam) botnets. A signifigant percentage (perhaps even a majority) of spam is sent this way. Of course, blocking port 25 is still retarded because all this spam actually just goes through the ISPs SMTP relay.
The belief that copyright law covers use is a particularly odious urban legend, and it's propagated by people with a vested interest who want you to believe that. It is *not true* - you do not need a copyright license to "use" copyrighted material. The rights reserved to copyright holders are explicitly enumerated, and "use" is not one of them. It is not possible, ever, for reading a book to be (in and of itself) a copyright violation. Similiarly, it is not possible to violate copyright simply by running software (although, of course, you may have violated copyright in order to aquire that software).
This is incorrect - mere "use" is not a right reserved for copyright holders. You don't need a license (implicit or explicit) to read a book, and reading a book cannot (in and of itself) be a copyright violation under any circumstances. Years go people tried to argue that copying to a hard drive (installation) and into RAM (use) were "copies" under copyright law and you therefore can't use software without a license, but thats been untrue since 1977, when "copying essential to use" was explicitly placed under fair use.
One, the GPL is a license, not a contract, and violations of it fall under (federal) copyright law, not contract law (and violation of the GPL could quite likely fall under criminal copyright infringment, although such a case has never been pressed). But thats not what he's talking about - he's talking about needing to report your IP ownership under Sarbanes-Oxely, and both failing to report that and lying in it are violations of (federal) securities law. So if you're violating the GPL (note that this doesn't cover normal use, but people who're distributing products based on Linux and the like), then your SOX statements will be either incomplete or false (or truthful, but that'd be stupid. I guess it's kinda like how your IRS tax forms say you have to declare embezzled or other illegal income), so it's not just some unwashed hippies suing you that you might have to worry about, but nasty feds coming down on you for SOX violations.
Wow, way to ruin the internet. This sort of thing is exactly what people are trying to fight - basing everything on the "community standards of the recipient" is a recipie for disaster when you're talking about global network (especially an anonymous, pull based one). If your law were passed, you'd have just given carte blanche to shut down almost any site in the US to *anyone* who can afford a plane ticket and the services of a 16 year old.
This was already used years ago to try to shut down the mail order porn industry - a DA would order something (via mail) to some county with a sympathetic judge and file suit there for violating community standards where it was recieved. It's an unacceptable burden to require someong fulfilling a request to first analyze the community standards of the reciepient, and the problem is even worse on the internet.
Lastly, it's important to remember that the internet is *not* like the real world, and that "community standards" a pretty questionable standard to apply to it anyway. Unlike physical locations, you can't be required to pass by a porn site in order to get to somewhere else. If you're looking at porn on the internet, then you're either doing it with full knowledge of your circumstances, someone has subverted your computer, or you're doing foolish image searches. And even if it's the last, I think it's extremely questionable that we need legislation to "protect" against this. I suspect that the amount of porn "delivered to children" when those children weren't actively seeking it out is extremely minimal and unlikely to happen enough to damage someone.
I'll give an allegory for the whole "accidental search" thing. When I was in high school a few friends and I were on a road trip to Seattle. We were wandering around the city and saw a sign for some shop that was something like "fantasy bookstore". I'm sure you can see where this is going - it was, of course, an adult sex toy/bookstore, not at all the right kind of fantasy. But just like when you mis-click on a search result, it took about 10 seconds for us to realize that we'd made a wrong turn and go back out. The fact that a minor can accidently walk through the door of an adult bookstore (much less a minor who actively tries to sneak in past the proprietor) does not mean we need legislation to "protect" that.
Morrowind is a good game, and a fun game, but it is not "hard". If you're a fan of Final Fantasy and other linear, stat based RPGs you won't have much fun, just like the OP. It's pretty easy, even at level 1, to steal very powerful equipment. If the fun for you in an RPG is leveling your character and improving the stats, then Morrowind is not for you - it is player and story driven. Exploring the hidden corners of Vvardenfell, discovering the secrets behind not only the main quest but the various side quests and tie ins is the enjoyment. It's not for everyone.
What the hell is wrong with people and the continual redefinition and apologetics for censorship? It's not censorship unless they shoot you for publishing it? Whats wrong with you?
This is a case of a government using it's authority to supress speech which they find objectionable. It is a textbook example of censorship and the fact that he can work around it by publishing in an area they don't control doesn't change that, any more than the fact that you could publish a book in the US that you couldn't in soviet Russia means that they didn't have censorship.
but who is to deny anyone, government controlled, influenced or not the right to protect their perceived "integrity".
There's this thing called free speech. A few people think it's important. Lots of other people wouldn't know it if it kicked them in the ass, and do stupid shit like nodding to each other about how sure, free speech is important, but you don't want anything bad out there, right?
Is this censorship? Hardly. We are not talking about some Great Firewall preventing the people from visiting any site of Cohen's.
Are you an idiot? Certainly. Any time *anyone* uses authority to suppress or alter someones speech, that is censorship. They don't have to kill you or prevent other people from listening or even be successfull. It's still censorship.
To say this is the beginning of state sponsored censorship is ridiculous
Yes, because it's not "the beginning" of anything. It *is* state sponsored censorship. It's what the word *means*.
Microsoft did a similiar thing a couple years ago, placing "stick on" MSN butterflies all over the sidewalks in Manhattan, except there they really were illegal cause they hadn't licensed anything from anyone. And there was almost exactly the same amount of distaste over the advertising.
MySQL claims that using MySQL via ODBC means you need to either abide by GPL or use a commercial license for MySQL (which I find questionable, but okay) but what I was talking about was the GPL "infecting" client software, not finalized solutions. It does *not* mean that any software anywhere that uses ODBC must be GPL because you might use it with MySQL.
Their legal principle is distribution rights of their software, not "derived works" rights over your software. Your are correct that the net-effect is the same, but the argument is quite different.
Untrue. The legal principle is use of the software, which is creating derived works (the only use for a library). Otherwise, for example, you could freely write programs using commercial third party libraries provided you didn't distribute them, which is not true.
The prohibition of this under copyright law has nothing to do with entity/employer/employee distinctions. Streaming the music involves making a copy, copyright law regulates the making of copies, and fair use/fair dealing does not provide an exception allowing copying in that case.
You are absolutely correct and this is why distribution of GPLed software within an organization should/would trigger the GPLs conditions. Distribution of software involves making a copy, copyright law regulates the making of copies, and fair use does not prove an exception allowing copying in that case. The GPL does, but only with certain limitations, including an offer to provide the source and not being able to restrict further distribution. Either distribution within a company is "copying" under copyright law or it is not - it's not one thing for GPLed software and another for everything else.
You've got an unsupported and unsupportable double standard here. When I get a company PC, and it has software loaded on it, that software is being *distributed* to me. If it's not distribution (and if it doesn't trigger GPL provisions, it cannot be distribution), then copyright law cannot apply, and while a software EULA may (questionably) apply, software or music with no EULA should be just as freely "distributable" as GPL software.
I don't know about turbulence, but planes have been (capable of) landing themselves on autopilot since the 70s. Taking off is harder but I believe autopilots can do that now as well. Autopilots today can also change course and altitude to avoid weather conditions - it's quite a bit more sophisticated than simply following a course. Driving on the ground is a much harder problem, but don't underestimate what autopilots are capable of.
It turns out that A and B are the same organization.
This is incorrect. B is an employee or other person within an organization. If your employer uses GPL-based software, and it is provided to you, you have the right to get the source for it, and you have the right to re-distribute it outside the organization. If every person within an organization was considered the same legal entity, then corporations would only need to purchase 1 license for every piece of software they used, one copy of every book they need, and it would be legal to set up music servers that streamed music off corporate-purchased CDs to everyone in the company without paying any further fees. Try doing any of those things and see how far you get.
Neither Oracle or MS claim any rights over my software program, but MySQL does.
It would be more correct to say that they do not *exercise* any rights over your software program. Re-read those license agreements. For your info, you do not write code that links against "the Oracle ODBC driver", but rather against the system ODBC libraries. Oracle (and Microsoft, for that matter) provide an ODBC driver that links into the the system ODBC and thats why your end user is responsible for aquiring the driver. Whoever writes the Oracle ODBC driver is responsible for abiding by the terms of the Oracle client library license, just like someone writing a MySQL ODBC library must abide by the terms of the MySQL license. Note that writing to an ODBC interface, and your clients using MySQL via it's (GPL) ODBC driver, does *not* require you to abide by the GPL.
However, if I understand the MySQL/FSF position correctly, I can't do the same thing with MySQL drivers because my application is considered to be a "derived work" of MySQL.
You can write code that works with MySQL via an ODBC interface and your only obligation is to whoever wrote your ODBC system libraries, but if you use the MySQL libraries directly (the so-called "native interface"), then your work is a derived work of the MySQL client libraries, which are GPLed. The *exact* same thing happens with the Oracle client libraries, except of course that the license is not GPL.
This may or may not be legally correct, but it is certainly a "gotcha" for folks used to commercial library licences.
I can't help but think that you don't actually work much with real honest to goodness commercial libraries and are really thinking of system libraries that ship with other products (like Windows). There are many commercial library vendors and they absolutely claim rights over your finished product - many of them require royalties.
The fact of the matter is that if you build an applicaiton that uses (say) the Oracle ODBC driver, you might owe Oracle some money to distribute it, but you don't have Oracle asserting intellectual property rights over the rest of your program. Which is exactly what MySQL is doing.
I'd suggest that you re-read the license agreement for the commercial libraries you use (including the Microsoft) ones, then. And *especially* re-read the licenses from the companies that sell third party libraries as thier business (you didn't just mean Microsoft when you said commercial vendor, right?). You can't use thier stuff without paying them. You can't use GPL stuff without "paying" in the form of obeying the license agreement. You can resolve both issues in the same way - by extracting all the original code from your program and shipping it seperately.
There is a principle called Abstraction Filtration Comparison Test, but you won't read about it on the FSF's website:)
Well, you will, but it's not relevant to this issue. The very first step of the test is to break the program into components - is a dynamically loaded library considered a component? If it is, then theres your infringment right there. How about statically linked? What about more dynamic solutions like runtime library loading? Furthermore, to my knowledge the abstraction/filtration comparison test has never been applied to an "executable", only to source code, but the courts have been quite clear that copyright does cover the binaries.
It's against the Windows HGI to do this now (I don't know if it was ever documented as the standard, but it was extremely common) although many companies (and installer frameworks) still do it. Windows HGI also says that if all you're installing is link(s) to your executable(s), you should put them directly in the start menu root instead of into a subfolder.
Re:Interestingly...
on
Why Use GTK+?
·
· Score: 2, Informative
The GPL implies that one can "use" the software without restrictions, and then relies on an extra-legal interpretation of copyright law to enforce the "viral-ness".
Not really. The viralness of the GPL is well founded - what you may be thinking of is how the FSF interperts "derived work", which is a legal grey area at the best of times. In the important cases, like when you statically link GPL code, it's pretty cut & dried. Dynamic linking is more complicated and as far as I know there's never been a court case that really spelled out the limits, and I doubt there would be - there still isn't a hard standard for music or written works. The main legal weakpoint of the GPL is actually in an area where it is overly permissive - distribution within an organization is absolutely considerd "distribution" under copyright law and should invoke the GPL, and I'm not sure why they think it shouldn't.
Not to mention that the FSF's take on the issue of libraries and derived works is entirely different than every commercial library vendor's!
This is totally false and you won't find any commercial library vendors saying that you don't owe them money because you only dynamically linked to thier library, or because you didn't distribute it.
The virtual printer technique won't preserve metadata because it's not document aware - it presents itself as a Windows printing device, the application uses standard Win API drawing commands on it, and the driver internally translates those commands into postscript and then into PDF. Do watch out for redactions, like drawing black boxes over text - the application will likely print this as the text with a box over it, which will look fine on paper but the redacted text will remain in the document. This happened at least once with a redatected DoD document, exposing them to some fairly serious embaressment because they're redacted all the negative parts of an independent audit.
The places you need to worry about metadata exposure are the document-aware "export" functionality, because rather than simply printing from primitives, these work with full knowledge of the document and it's structure.
Gah. Did you even bother to fucking check before spouting this traditional "can't rely on OSS" crap? MySQL AB is a commercial company that owns the copyright to MySQL and employs every major MySQL developer. They make money off support contracts and non-GPL licensed versions of MySQL. I can't speak for how *good* thier support is, as I've never used it, but apparently neither can you.
And I'm an avowed MySQL hater. I think it's a shitty hack of a database and there is no problem domain where there isn't at least one other product is a better solution. But this "no support" line is almost as wrong as it is stupid.
By the way, the kind of support you're claiming you need (24/7 on call support with access to someone who can provide a patch for a new problem) will be enormously expensive, where it's even available - there's no support plan Microsoft provides that makes those sort of promises, for example. You might get a patch in a half hour if it's a known problem and they have one ready for it, but you'll probably spend more than that half hour just convincing the rep that you have that specific problem. IBM or some of the other more service oriented companies may provide that kind of support. It's going to be expensive though - expensive enough that if you're worried about how much it's going to cost, it's probaby more than your company is worth.
Enterprise software is very rarely sold shrink wrapped with an EULA.
True, but what is being compared is the freely available "express" versions of these enterprise databases, and are covered by your traditional clickwrap EULA (much stronger than a retail shrinkwrap one, by the way), not by the (yet even stronger) negotiated enterprise contract.
If you're interested in typing in Python, you may be interested in looking at the PyPy project, which has done a ton of cool work (which, in the best tradition of academics everywhere, isn't really usable for anything pratical yet:P), one of which is a type inferrence engine which works on a (limited, hardly any dynamism) subset of Python.
I honestly wasn't trying to open the whole static/dynamic flamewar bag, although I realized that I went down that path without thinking. The issue I was raising is that, when you're talking about bulletproofing, static typing is insufficent - it obviously does catch classes of errors (at compile time) that dynamic typing cannot. However, the claim is generally extended that this makes it the only thing suitable for "mission critical" systems, which I consider false - if it's "mission critical" and needs to be proven, complete test coverage will work as a superset of the safety static typing can give you. Of course, you then need to have complete test coverage - but that is one of the costs of "mission critical" systems. Proponents of dynamic typing (can you guess which side I'm on?:P) will say that since 100% test coverage provides superior safety to static type checking, your time is better spent figuring out wasy to make test coverage better and running tests easier than arguing with the compiler about whether your program is runnable.
And, of course, the term "mission critical" is way overused. I know of literally mission critical systems, like the company goes broke and we all are out of work critical, that run on Access databases on world-writable shared drives and rely on the "yell over the cube wall" method of file locking.
(I know I responded to the parent, but you were agreeing with him anyway:P)
Static typing, the concept, provides a degree of provability, but every popular language that is statically typed has commonly used methods for escaping and subverting the type system which render proofs of the type system useless unless you avoid those features. I'm not even sure that it's possible to write a non-trivial Java application that way (prior to Java 5) unless you re-implement much of the standard library, and I'm not sure I've seen even trivial C applications without casts in them. So my submission is that static typing, as implemented in Java/C#/C++/C (the 4 statically typed languages I would consider popular), cannot provide you with any meaningful provability. You can get better provability from Haskel or some other functional languages (type inference is very powerful), but even then, proving the type system isn't sufficent to guarantee the programs behavior. You can get that only through testing that would catch type errors anyway. It's the difference between provably sound (in a mathematical sense), and proven to work (in a real world sense). The latter simply can't be done - you can only prove that it's worked "so far".
Anti-Java hostility is common in Pythonistas and whatever Rubists call themselves. It's like you met this woman at a nice, upscale party (hosted by Sun, of course), and so you dated her for a while and she was pretty nice - not great, not especially good looking, but you got along okay and you aren't a prize yourself. So everything was okay for a while but then you noticed a few things about her. Like that she took forever to get up in the morning, and that she made you move to her house instead of living in yours, and she never let you hang out with your friends, and although she claimed to be a neat freak and hate garbage, she had a tendency to refuse to throw anything out because she thought she might need it later. And then the real bad stuff started to happen, like overnight she tripled in size and you don't know why, and she started insisting that you write everything down in Spanish before translating it to English to tell her. And then she killed your dog, and you were scared to leave.
So one night after a long day satisfying your demanding new mistress, you run down to a bar you used to hang out with some friends, nad you meet this new chick. And she's really really pretty - elegant, even, and a hell of a dancer. And she thinks just like you do, you can talk about anything and you end up completing each others sentences before the night is over. She's a part of your crowd - all your friends know her and hang out with her too, no jealousy involved. She has her own house but has no problem living in yours, doesn't spread her stuff out too much, picks up after herself (but sometimes you have to remind her). So you ditch the demanding bitch and you move in with the new girl. And whenever you hear someone talking about those upscale women you meet at the big parties, you just get a bad taste in your mouth, and you tell them about the hot little lady you met at the bar.
Python is as suitable as any other (popular) language for mission-critical applications, including Java. The kinds of things you need to do to ensure reliability to a "mission critical" level - a quality which is not easily defined and more honored in the breach than in the observation, I might add - have very little to do with programming languages themselves, and to the extent that you can implement them in the language itself, no popular language does so. Static typing is insufficent, and the measures you would have to take would superscede any static typing you do. Of course, there is always interest in getting better, even if you can't get perfect. My experience is that Python code tends to be more reliable under optimal conditions because given the same time to completion, you can spend more time testing and exercising it (the *only* way to know if a program is reliable and stable is to test it, and even then you only know up to the limits of your testing)
You can quickly create a simple program but it is hard to be sure that it is really stable
If you used the same methodology on a Java program and a Python program, then you can be equally sure. If you didn't exhaustively test your Java program, then you aren't any more sure about it (or you shouldn't be) than you are about your Python one.
Slashdot Mirror
No, absolutely wrong. Contract law is a seperate category and is not related to copyright law. A copyright license is different legal entity than a contract, and violating one is not a breach of contract - it is a totally seperate action, falling under different legislation, with different issues and different penalties.
Second, since you no longer have permission to use the copy of the software in your possession, you are in violation of federal copyright law
Blah blah blah, copyright doesn't cover use, blah blah blah.
In the portion that I quoted, he vaguely referenced "federal law" for violating a license.
If you'd use that magic scrollbar thingy and look down a couple paragraphs you'd see where he mentions securities law. Maybe you should read the *whole* article before posting next time. Of course, even if you don't know how to work a scrollbar, theres this thing called "context" that might lead you to believe he's talking about Sarbanes-Oxley.
Next, I discussed how the Sarbanes-Oxley provision might apply, and why I didn't think that it did.
No, you made up some random case that wouldn't be covered by SOX and knocked down the strawman iwth great strength and power, totally ignoring the actual point made by the article.
It's not spammers, it's compromised machines used as (spam) botnets. A signifigant percentage (perhaps even a majority) of spam is sent this way. Of course, blocking port 25 is still retarded because all this spam actually just goes through the ISPs SMTP relay.
The belief that copyright law covers use is a particularly odious urban legend, and it's propagated by people with a vested interest who want you to believe that. It is *not true* - you do not need a copyright license to "use" copyrighted material. The rights reserved to copyright holders are explicitly enumerated, and "use" is not one of them. It is not possible, ever, for reading a book to be (in and of itself) a copyright violation. Similiarly, it is not possible to violate copyright simply by running software (although, of course, you may have violated copyright in order to aquire that software).
This is incorrect - mere "use" is not a right reserved for copyright holders. You don't need a license (implicit or explicit) to read a book, and reading a book cannot (in and of itself) be a copyright violation under any circumstances. Years go people tried to argue that copying to a hard drive (installation) and into RAM (use) were "copies" under copyright law and you therefore can't use software without a license, but thats been untrue since 1977, when "copying essential to use" was explicitly placed under fair use.
One, the GPL is a license, not a contract, and violations of it fall under (federal) copyright law, not contract law (and violation of the GPL could quite likely fall under criminal copyright infringment, although such a case has never been pressed). But thats not what he's talking about - he's talking about needing to report your IP ownership under Sarbanes-Oxely, and both failing to report that and lying in it are violations of (federal) securities law. So if you're violating the GPL (note that this doesn't cover normal use, but people who're distributing products based on Linux and the like), then your SOX statements will be either incomplete or false (or truthful, but that'd be stupid. I guess it's kinda like how your IRS tax forms say you have to declare embezzled or other illegal income), so it's not just some unwashed hippies suing you that you might have to worry about, but nasty feds coming down on you for SOX violations.
This was already used years ago to try to shut down the mail order porn industry - a DA would order something (via mail) to some county with a sympathetic judge and file suit there for violating community standards where it was recieved. It's an unacceptable burden to require someong fulfilling a request to first analyze the community standards of the reciepient, and the problem is even worse on the internet.
Lastly, it's important to remember that the internet is *not* like the real world, and that "community standards" a pretty questionable standard to apply to it anyway. Unlike physical locations, you can't be required to pass by a porn site in order to get to somewhere else. If you're looking at porn on the internet, then you're either doing it with full knowledge of your circumstances, someone has subverted your computer, or you're doing foolish image searches. And even if it's the last, I think it's extremely questionable that we need legislation to "protect" against this. I suspect that the amount of porn "delivered to children" when those children weren't actively seeking it out is extremely minimal and unlikely to happen enough to damage someone.
I'll give an allegory for the whole "accidental search" thing. When I was in high school a few friends and I were on a road trip to Seattle. We were wandering around the city and saw a sign for some shop that was something like "fantasy bookstore". I'm sure you can see where this is going - it was, of course, an adult sex toy/bookstore, not at all the right kind of fantasy. But just like when you mis-click on a search result, it took about 10 seconds for us to realize that we'd made a wrong turn and go back out. The fact that a minor can accidently walk through the door of an adult bookstore (much less a minor who actively tries to sneak in past the proprietor) does not mean we need legislation to "protect" that.
Morrowind is a good game, and a fun game, but it is not "hard". If you're a fan of Final Fantasy and other linear, stat based RPGs you won't have much fun, just like the OP. It's pretty easy, even at level 1, to steal very powerful equipment. If the fun for you in an RPG is leveling your character and improving the stats, then Morrowind is not for you - it is player and story driven. Exploring the hidden corners of Vvardenfell, discovering the secrets behind not only the main quest but the various side quests and tie ins is the enjoyment. It's not for everyone.
This is a case of a government using it's authority to supress speech which they find objectionable. It is a textbook example of censorship and the fact that he can work around it by publishing in an area they don't control doesn't change that, any more than the fact that you could publish a book in the US that you couldn't in soviet Russia means that they didn't have censorship.
but who is to deny anyone, government controlled, influenced or not the right to protect their perceived "integrity".
There's this thing called free speech. A few people think it's important. Lots of other people wouldn't know it if it kicked them in the ass, and do stupid shit like nodding to each other about how sure, free speech is important, but you don't want anything bad out there, right?
Is this censorship? Hardly. We are not talking about some Great Firewall preventing the people from visiting any site of Cohen's.
Are you an idiot? Certainly. Any time *anyone* uses authority to suppress or alter someones speech, that is censorship. They don't have to kill you or prevent other people from listening or even be successfull. It's still censorship.
To say this is the beginning of state sponsored censorship is ridiculous
Yes, because it's not "the beginning" of anything. It *is* state sponsored censorship. It's what the word *means*.
Microsoft did a similiar thing a couple years ago, placing "stick on" MSN butterflies all over the sidewalks in Manhattan, except there they really were illegal cause they hadn't licensed anything from anyone. And there was almost exactly the same amount of distaste over the advertising.
MySQL claims that using MySQL via ODBC means you need to either abide by GPL or use a commercial license for MySQL (which I find questionable, but okay) but what I was talking about was the GPL "infecting" client software, not finalized solutions. It does *not* mean that any software anywhere that uses ODBC must be GPL because you might use it with MySQL.
Their legal principle is distribution rights of their software, not "derived works" rights over your software. Your are correct that the net-effect is the same, but the argument is quite different.
Untrue. The legal principle is use of the software, which is creating derived works (the only use for a library). Otherwise, for example, you could freely write programs using commercial third party libraries provided you didn't distribute them, which is not true.
You are absolutely correct and this is why distribution of GPLed software within an organization should/would trigger the GPLs conditions. Distribution of software involves making a copy, copyright law regulates the making of copies, and fair use does not prove an exception allowing copying in that case. The GPL does, but only with certain limitations, including an offer to provide the source and not being able to restrict further distribution. Either distribution within a company is "copying" under copyright law or it is not - it's not one thing for GPLed software and another for everything else.
You've got an unsupported and unsupportable double standard here. When I get a company PC, and it has software loaded on it, that software is being *distributed* to me. If it's not distribution (and if it doesn't trigger GPL provisions, it cannot be distribution), then copyright law cannot apply, and while a software EULA may (questionably) apply, software or music with no EULA should be just as freely "distributable" as GPL software.
I don't know about turbulence, but planes have been (capable of) landing themselves on autopilot since the 70s. Taking off is harder but I believe autopilots can do that now as well. Autopilots today can also change course and altitude to avoid weather conditions - it's quite a bit more sophisticated than simply following a course. Driving on the ground is a much harder problem, but don't underestimate what autopilots are capable of.
This is incorrect. B is an employee or other person within an organization. If your employer uses GPL-based software, and it is provided to you, you have the right to get the source for it, and you have the right to re-distribute it outside the organization. If every person within an organization was considered the same legal entity, then corporations would only need to purchase 1 license for every piece of software they used, one copy of every book they need, and it would be legal to set up music servers that streamed music off corporate-purchased CDs to everyone in the company without paying any further fees. Try doing any of those things and see how far you get.
It would be more correct to say that they do not *exercise* any rights over your software program. Re-read those license agreements. For your info, you do not write code that links against "the Oracle ODBC driver", but rather against the system ODBC libraries. Oracle (and Microsoft, for that matter) provide an ODBC driver that links into the the system ODBC and thats why your end user is responsible for aquiring the driver. Whoever writes the Oracle ODBC driver is responsible for abiding by the terms of the Oracle client library license, just like someone writing a MySQL ODBC library must abide by the terms of the MySQL license. Note that writing to an ODBC interface, and your clients using MySQL via it's (GPL) ODBC driver, does *not* require you to abide by the GPL.
However, if I understand the MySQL/FSF position correctly, I can't do the same thing with MySQL drivers because my application is considered to be a "derived work" of MySQL.
You can write code that works with MySQL via an ODBC interface and your only obligation is to whoever wrote your ODBC system libraries, but if you use the MySQL libraries directly (the so-called "native interface"), then your work is a derived work of the MySQL client libraries, which are GPLed. The *exact* same thing happens with the Oracle client libraries, except of course that the license is not GPL.
This may or may not be legally correct, but it is certainly a "gotcha" for folks used to commercial library licences.
I can't help but think that you don't actually work much with real honest to goodness commercial libraries and are really thinking of system libraries that ship with other products (like Windows). There are many commercial library vendors and they absolutely claim rights over your finished product - many of them require royalties.
Windows HGI says not to put links to your uninstaller in the start menu too :P
I'd suggest that you re-read the license agreement for the commercial libraries you use (including the Microsoft) ones, then. And *especially* re-read the licenses from the companies that sell third party libraries as thier business (you didn't just mean Microsoft when you said commercial vendor, right?). You can't use thier stuff without paying them. You can't use GPL stuff without "paying" in the form of obeying the license agreement. You can resolve both issues in the same way - by extracting all the original code from your program and shipping it seperately.
There is a principle called Abstraction Filtration Comparison Test, but you won't read about it on the FSF's website :)
Well, you will, but it's not relevant to this issue. The very first step of the test is to break the program into components - is a dynamically loaded library considered a component? If it is, then theres your infringment right there. How about statically linked? What about more dynamic solutions like runtime library loading? Furthermore, to my knowledge the abstraction/filtration comparison test has never been applied to an "executable", only to source code, but the courts have been quite clear that copyright does cover the binaries.
It's against the Windows HGI to do this now (I don't know if it was ever documented as the standard, but it was extremely common) although many companies (and installer frameworks) still do it. Windows HGI also says that if all you're installing is link(s) to your executable(s), you should put them directly in the start menu root instead of into a subfolder.
Not really. The viralness of the GPL is well founded - what you may be thinking of is how the FSF interperts "derived work", which is a legal grey area at the best of times. In the important cases, like when you statically link GPL code, it's pretty cut & dried. Dynamic linking is more complicated and as far as I know there's never been a court case that really spelled out the limits, and I doubt there would be - there still isn't a hard standard for music or written works. The main legal weakpoint of the GPL is actually in an area where it is overly permissive - distribution within an organization is absolutely considerd "distribution" under copyright law and should invoke the GPL, and I'm not sure why they think it shouldn't.
Not to mention that the FSF's take on the issue of libraries and derived works is entirely different than every commercial library vendor's!
This is totally false and you won't find any commercial library vendors saying that you don't owe them money because you only dynamically linked to thier library, or because you didn't distribute it.
The places you need to worry about metadata exposure are the document-aware "export" functionality, because rather than simply printing from primitives, these work with full knowledge of the document and it's structure.
And I'm an avowed MySQL hater. I think it's a shitty hack of a database and there is no problem domain where there isn't at least one other product is a better solution. But this "no support" line is almost as wrong as it is stupid.
By the way, the kind of support you're claiming you need (24/7 on call support with access to someone who can provide a patch for a new problem) will be enormously expensive, where it's even available - there's no support plan Microsoft provides that makes those sort of promises, for example. You might get a patch in a half hour if it's a known problem and they have one ready for it, but you'll probably spend more than that half hour just convincing the rep that you have that specific problem. IBM or some of the other more service oriented companies may provide that kind of support. It's going to be expensive though - expensive enough that if you're worried about how much it's going to cost, it's probaby more than your company is worth.
True, but what is being compared is the freely available "express" versions of these enterprise databases, and are covered by your traditional clickwrap EULA (much stronger than a retail shrinkwrap one, by the way), not by the (yet even stronger) negotiated enterprise contract.
I honestly wasn't trying to open the whole static/dynamic flamewar bag, although I realized that I went down that path without thinking. The issue I was raising is that, when you're talking about bulletproofing, static typing is insufficent - it obviously does catch classes of errors (at compile time) that dynamic typing cannot. However, the claim is generally extended that this makes it the only thing suitable for "mission critical" systems, which I consider false - if it's "mission critical" and needs to be proven, complete test coverage will work as a superset of the safety static typing can give you. Of course, you then need to have complete test coverage - but that is one of the costs of "mission critical" systems. Proponents of dynamic typing (can you guess which side I'm on? :P) will say that since 100% test coverage provides superior safety to static type checking, your time is better spent figuring out wasy to make test coverage better and running tests easier than arguing with the compiler about whether your program is runnable.
And, of course, the term "mission critical" is way overused. I know of literally mission critical systems, like the company goes broke and we all are out of work critical, that run on Access databases on world-writable shared drives and rely on the "yell over the cube wall" method of file locking.
Static typing, the concept, provides a degree of provability, but every popular language that is statically typed has commonly used methods for escaping and subverting the type system which render proofs of the type system useless unless you avoid those features. I'm not even sure that it's possible to write a non-trivial Java application that way (prior to Java 5) unless you re-implement much of the standard library, and I'm not sure I've seen even trivial C applications without casts in them. So my submission is that static typing, as implemented in Java/C#/C++/C (the 4 statically typed languages I would consider popular), cannot provide you with any meaningful provability. You can get better provability from Haskel or some other functional languages (type inference is very powerful), but even then, proving the type system isn't sufficent to guarantee the programs behavior. You can get that only through testing that would catch type errors anyway. It's the difference between provably sound (in a mathematical sense), and proven to work (in a real world sense). The latter simply can't be done - you can only prove that it's worked "so far".
Anti-Java hostility is common in Pythonistas and whatever Rubists call themselves. It's like you met this woman at a nice, upscale party (hosted by Sun, of course), and so you dated her for a while and she was pretty nice - not great, not especially good looking, but you got along okay and you aren't a prize yourself. So everything was okay for a while but then you noticed a few things about her. Like that she took forever to get up in the morning, and that she made you move to her house instead of living in yours, and she never let you hang out with your friends, and although she claimed to be a neat freak and hate garbage, she had a tendency to refuse to throw anything out because she thought she might need it later. And then the real bad stuff started to happen, like overnight she tripled in size and you don't know why, and she started insisting that you write everything down in Spanish before translating it to English to tell her. And then she killed your dog, and you were scared to leave.
So one night after a long day satisfying your demanding new mistress, you run down to a bar you used to hang out with some friends, nad you meet this new chick. And she's really really pretty - elegant, even, and a hell of a dancer. And she thinks just like you do, you can talk about anything and you end up completing each others sentences before the night is over. She's a part of your crowd - all your friends know her and hang out with her too, no jealousy involved. She has her own house but has no problem living in yours, doesn't spread her stuff out too much, picks up after herself (but sometimes you have to remind her). So you ditch the demanding bitch and you move in with the new girl. And whenever you hear someone talking about those upscale women you meet at the big parties, you just get a bad taste in your mouth, and you tell them about the hot little lady you met at the bar.
You can quickly create a simple program but it is hard to be sure that it is really stable
If you used the same methodology on a Java program and a Python program, then you can be equally sure. If you didn't exhaustively test your Java program, then you aren't any more sure about it (or you shouldn't be) than you are about your Python one.