I did RTFA. And yes, the law was intended to work that way. Unfortunately, that's not what the law says. And since almost any work on a computer involves investigating data on that computer not accessible to the public (the user's firewall settings, for example, aren't available to the public), any such work falls under the "investigation" part and requires a PI license.
And the law will be enforced based on what it says, not on what anyone thinks it should have said instead.
It may be a reasonable position, but is it a correct one? So far nobody's been able to come up with any hard numbers agreeing with it. All the results are either "no correlation" or "kids who play violent video games are slightly less likely to be violent than average".
The position that if you're standing in the airlock of the ISS and give a good shove off, sending yourself flying away from the station and towards Earth, you'll burn up in the atmosphere is also a reasonable one. It just happens not to be correct. One orbit later you'll find yourself bumping off the ISS again. Common sense might say one thing, but orbital mechanics says another. I suspect the same thing's at work: common sense might say that violent games should beget violent acts, but reality and psychology don't work the way common sense says they should. Nothing new there, lots of things that're true violate common sense.
True, you can find out that information easily. What boggles me is that that information could pose a threat for identity theft. It's publicly-known information. To impersonate someone well enough to get credit or open a bank account or anything in their name, publicly-known information should not be sufficient.
But what's particularly sensitive about any of that information you listed? Name, address and phone number, well, I'm in the phone book, anybody who knows my name can get that information in a few seconds. Who I work for? That's not particularly sensitive (unless you work for No Such Agency or something), and anybody who knows your address can probably ask around and find out if they don't already know. Age and date of birth? Nobody's particularly careful about their age, and date of birth's required so many places that it has to be assumed to be widely known. If anybody knows that, all it shows is that they know who I am. If any entity assumes that knowledge of that information implies the person who knows it is me, that entity obviously has trouble getting two neurons firing in sync.
Also in this case probably a violation of the Video Privacy Protection Act. YouTube sounds like they qualify as a "video tape service provider" in (a)(4) ("delivery", "similar audio visual materials"). (b)(2)(F) covers disclosure in a civil proceeding, Viacom hasn't given prior notice to consumers of their request for disclosure and YouTube users haven't been given the opportunity to appear to contest Viacom's request, both are required by that paragraph (the following part about appropriate safeguards applies only if the order's granted pursuant to (F), and since the requirements of (F) weren't met the order couldn't have been properly issued pursuant to it).
There's actually a solution to that, but it involves slowing the process down. Just don't activate the account once the information's entered. Instead, send a physical letter to the credit-card billing address. You can require a form to be signed and returned, or just include an activation code in the letter that has to be entered to turn the account on. That should make it infeasible to use 99% of stolen cards. It introduces a few days of delay between requesting the account and getting it, but IMO if you intend to use the account for any length of time a few days shouldn't be an issue and if you don't then you're likely exactly the kind of person this is intended to filter out.
Amazon has billing information for those accounts. Money changes hands. So, require that the name and address given to Amazon when setting up the account match the billing name and address of the credit card used to pay for the services. Most mail-order and on-line merchants do that already, and won't ship except to the billing name/address. Then block known pre-paid debit card numbers and one-time card numbers. Not perfect, but it should knock down 90% of the problems and make it a lot harder for a criminal to get a new account not tied to a consistent identity.
Re:GPLv2 and GPLv3 have the same spirit
on
A Year of GPLv3
·
· Score: 1
If I wanted to keep my code private, then I wouldn't use GPL'd code as the basis for my firmware, simple as that. When I use someone else's code, I have to abide by their license terms. If I can't or won't do so, I don't use their code. Nothing specific to GPL'd software there, the same thing applies to commercial software. If I don't want to or can't pay the royalties MS requires for Windows CE, I don't use Windows CE in my devices.
If your only complain is "I can't use someone else's software without complying with their terms!", then I think you'll get precious little sympathy here.
I think the key question would be this: can you modify the code on the board without a dev kit? If you've got some way of getting your code onto the board without a dev kit, or by using some piece of hardware or software the user's prohibited from obtaining, then you may run afoul of the GPLv3's terms. But if all you need's a dev kit, both you and the user would need that dev kit to change the software and you aren't prohibiting the user from purchasing that dev kit, then you'll be OK. GPLv3 doesn't require that you let the user modify the code running on the device, just that you not block them from access while retaining that access yourself. As someone else noted, if you burned the code into ROM then the user can't modify it, but that's acceptable under the GPLv3 because you can't modify it either (short of burning a new ROM, going out and physically installing it, which the user could do too if they were so inclined and capable).
Just bear in mind that this'll prevent proper handling of hosts that use tcp_wrappers and the inetd superserver to handle services and access control. That combination works by accepting the connection, getting the peer address (must be connected for this), checking the access rules and immediately issuing an RST before any data's been transferred if access is denied. This rule prevents that RST from getting through, leaving you handing trying to connect to a port that's not going to allow you to connect.
And so it comes full circle. The ancient flat hosts file that the ARPAnet used way back in it's early days, the one that was abandoned in favor of hierarchical DNS because it wasn't possible to manage a flat namespace when the net was a few thousand machines, returns. What, you think the companies that insist on registering every variation of their name in every domain in existence won't insist on having their own TLD too? And we'll be back to hostnames being of the form "ford". And "fordfocus". And "focus". And "myfocus". And "myford". All belonging to Ford Motors. And the inevitable fights when Focus Magazine (a fine-art photography magazine) also wants "focus" and has the trademark to justify getting it just as much as Ford.
And as any lawyer, or any businessman, will tell you, when the other party says "There's no problem, we don't need to put that in writing.", you will be needing it in writing at some point and you'll be regretting not getting it.
LimitNone's only hope is language in a written contract promising that Google won't compete with their product. Absent that, they're toast. Let's face it, their product isn't much. There's nothing in it that hasn't been well-known and in wide use for the last 30 years, and thus can't be a trade secret. Google obviously knows their own formats and APIs for loading messages into gMail. In fact they had to have created them before LimitNone's product existed, if they hadn't LN couldn't have created their product at all. The file formats and things like Exchange server APIs aren't exactly secret. Copying all messages from one to the other is the same basic copy loop that's been used for 40-some years: open input; while not eof(input) do read(input); write(output); done; close input. A loop to iterate through folders and some recursion to handle subfolders, I was doing that in high school. Look and feel? LimitNone's probably using the standard tree-view widgets provided by the system, so yes Google's app will look like theirs because both of them look like the standard system widgets. That's assuming the apps allow message-level selectivity, if they limit it to folder-level or "everything on this server", the UI's even more generic. And the concept of importing mail messages from an old client into a new one? Hardly new. Mail clients have been importing other client's mailboxes since as long as there've been mail clients. Thunderbird has been doing it IIRC for a couple years now, well before LimitNone's product was created.
LimitNone's problem is that they're trying to charge $29 for a basic one-shot function that comes standard with most mail clients and that frankly could be hacked up by a single programmer in a few weeks of full-time work.
Yes, but that's the point: everybody does know (or any reasonable person should know) that that happens. Servers cease to exist, servers move to other names/locations, the people running the server decide they don't want to host older versions anymore when newer ones are available. And while the courts will tend to do what you suggest when someone couldn't reasonably have known, they're less tolerant of people who knew full well what would happen and did it anyway.
As far as being nice to companies, I ask this: if you broke their license knowing full well what you were doing, how would they treat you? And why should they expect to be treated any differently than they'd treat someone in their position? That's my touchstone: I'll cut them exactly as much slack as they'd cut anyone else. Which is usually precious little. And if I were running a business, I'd simply avoid the whole problem by either complying with the license or not distributing that software. Then the problem never comes up and I don't have to worry about whether anybody'll cut me any slack.
Well, the BusyBox people aren't bound by the GPL's terms. They're the copyright owners, they don't need a license to distribute their own works. They could stop tomorrow, take everything down, close up shop without any further legal obligations to anyone. Only the people who redistributed their code are under any legal obligation to continue providing the source code.
And GPL v2 is even worse. Under it offering access from a network server is not sufficient to satisfy it's section 3b requirements. If you distribute via section 3a and distribute only by offering download from a server then you can offer the sources from the same server and location as the binaries and be OK, but 3b doesn't allow that option in GPL v2. If someone asks, you have to provide the source on a distribution medium. Telling them to go to someone else isn't providing the source code to them on a distribution medium, I'm afraid.
Except for one small detail: the GPL doesn't require that the source code be available. It requires that the party doing the distributing (or redistributing) make the source code available. Read section 3 of the GPL v2 or section 6 of the GPL v3. The clearest statement is in GPL v3 section 6d, where it says clearly "Regardless of what server hosts the corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.". The servers hosting the BusyBox code aren't obliged to keep version 1.n.n up, and if you depend on them and they take it down you are the one in violation of the GPL by not being able to provide access to the source code as required.
You're correct in thinking that. You don't have to provide access to just any source code for what you distributed, you have to provide source code for the version you distributed. If you don't have an agreement with the party hosting the source code you point to, they may update to a newer version or cease hosting it completely without any warning to you. And since you are the one obligated to provide the source code or access to it, not them, you'll be the one on the hook for copyright violation if/when that happens. And I'd note that it likely wouldn't be considered "unwitting" if the case went to court, since any reasonable person would know or should have known that this would occur.
IANAL, but every time I've talked to one "distribution" depends on who it's being distributed to. If I'm at both ends of the transaction, it is not distribution in the sense the GPL uses, any more than it's distribution in the GPL sense if I copy the code from my desktop box to the server box sitting next to it and plugged into the same Ethernet switch just above them. Physical distance doesn't matter, it's ownership/control of the ends and if I own or have sole control over both ends then it's just me and there's no distribution involved in the legal sense. Legal here is just being paranoid, or they're somewhat clueless about networks and software (not a good thing for the lawyers of a company that deals with both, IMO).
Now, if the servers aren't under your sole control, if it's a shared system where other entities than your company have root access and/or significant control over the systems without your supervision, that would be a different case.
Except that some people don't realize what's happening. They put a file into their shared folder so they can copy it to another of their computers on the network, not realizing that in the process it's being made visible to everyone else who looks.
Analogous situation: I like to read out on the porch, so I've got a stack of books on the table on the porch so they're conveniently reachable. I hadn't realized that someone else had left the gate open so anybody wandering by could walk in and make a copy of any book they wanted and walk away with it. Certainly there'd be no point in my making the books available to everybody else, except that I didn't realize I was making them available. And gods know the majority of people are computer-clueless and have no idea what a "shared folder" is, they just think "No, that isn't a shared folder, that's just the folder I put things in when I need to copy them to another computer. I'm not sharing with anybody, it's all my computers.".
Possibly you're right. However, I'm minded of something that happened when I was living in rural Nevada. One of the (newish) salesmen for the Ford dealer saw a cowboy wandering around looking at the luxury cars on the lot. The cowboy was a real rough-looking type, dirty, sweaty, holes in his jeans, looked about 60 and was driving a half-rusted-out pickup. The salesman ignored him, and after a while the cowboy wandered over and asked about one of the more expensive cars. The salesman blew him off, figured this guy wasn't going to be able to afford that car, not as poor as he looked. The cowboy left.
The next day, the dealership owner calls everybody in. He hauls the salesman above up in front of everybody and chews him out royally, ending with "You're fired! Get off my lot, NOW!". It turns out that one of the biggest ranchers in the area, a guy who doesn't bother writing checks for less than 10 grand because he's got that much in cash in his pocket on any given day, who'd been buying all his ranch's vehicles from this dealership for years, just called in to say he was ending all his leases and taking his business down the road to the Dodge dealer. That added up to a lot of sales lost. Apparently the rancher had been in the day before looking for a new car, with cash in his pocket ready to buy on the spot, and the salesman wouldn't even give him the time of day.
When you're deciding that it's not worth your time to pay attention to something, be very very sure you've taken everything into account.
f, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Program at all. For example, if a patent
license would not permit royalty-free redistribution of the Program by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Program.
If you can't for some reason make available all the tools needed to build your derivative work, and those tools are not standard parts of a system or otherwise available independently, then you won't be able to comply with the GPL and thus won't be able to distribute while you're required to comply with the GPL. However, if you were using only the standard Microsoft compilers and such, they'd probably fall into the "standard system software" category just like GCC does (one must distribute the makefiles needed to build a GPL'd work, but one doesn't need to distribute GCC and make along with it).
Because it isn't their application. It's the source to the BusyBox author's application, which they're distributing without permission to do so (they aren't complying with the terms of the only license which would give them permission). I think it ought to come as no surprise to any sensible adult that you don't get to do anything you like with someone else's stuff.
Odds on the developers understand the GPL and it's obligations completely. But the packaging and distribution of the product isn't handled by the developers, it's handled by Marketing and Sales. And those guys probably don't even know the software has a license attached, they've got no clue what all went into their software, and they likely don't think they ought to consult with mere software developers about how they can market the product. And they probably didn't ask Legal for an opinion, since it's "their" software and they can (in Marketing's world) do whatever they want with it. Customer Support's likely a division of Marketing, so when the initial e-mails came in they got handled by people with that attitude.
And then one day the package with the letter on a lawyer's letterhead with all the court paperwork arrived, and it went straight to Legal, bypassing Marketing entirely. And Legal, being sensible people, asked the obvious questions of the obvious people. Marketing may ignore the engineers, but Management tends to listen to the lawyers.
Almost no one out side the free software community uses that definition of "free"? Oh, so when the star of The Prisoner said "I am not a number! I am a free man!", he meant he didn't cost any money? Oh, oops, no he didn't. The definition the free software community uses is, I'm afraid, one of the two common definitions for the word "free" that are in common use today. That you'd rather it only had one meaning, and that the monetary one, doesn't change the dictionary.
Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange;
See the bolded words? They don't say "any direct recipient" or "anyone who got the binaries from you", they say "any third party". Remember that the people who received binaries from you may have redistributed them under section 3c, in which case those people who didn't get the binaries directly from you are still entitled under the GPL to receive source code from you (by taking advantage of the written offer the original recipient passed along to them).
We've covered this time and time again. Why do people keep insisting that the GPL doesn't say what it says?
Slashdot Mirror
I did RTFA. And yes, the law was intended to work that way. Unfortunately, that's not what the law says. And since almost any work on a computer involves investigating data on that computer not accessible to the public (the user's firewall settings, for example, aren't available to the public), any such work falls under the "investigation" part and requires a PI license.
And the law will be enforced based on what it says, not on what anyone thinks it should have said instead.
It may be a reasonable position, but is it a correct one? So far nobody's been able to come up with any hard numbers agreeing with it. All the results are either "no correlation" or "kids who play violent video games are slightly less likely to be violent than average".
The position that if you're standing in the airlock of the ISS and give a good shove off, sending yourself flying away from the station and towards Earth, you'll burn up in the atmosphere is also a reasonable one. It just happens not to be correct. One orbit later you'll find yourself bumping off the ISS again. Common sense might say one thing, but orbital mechanics says another. I suspect the same thing's at work: common sense might say that violent games should beget violent acts, but reality and psychology don't work the way common sense says they should. Nothing new there, lots of things that're true violate common sense.
True, you can find out that information easily. What boggles me is that that information could pose a threat for identity theft. It's publicly-known information. To impersonate someone well enough to get credit or open a bank account or anything in their name, publicly-known information should not be sufficient.
But what's particularly sensitive about any of that information you listed? Name, address and phone number, well, I'm in the phone book, anybody who knows my name can get that information in a few seconds. Who I work for? That's not particularly sensitive (unless you work for No Such Agency or something), and anybody who knows your address can probably ask around and find out if they don't already know. Age and date of birth? Nobody's particularly careful about their age, and date of birth's required so many places that it has to be assumed to be widely known. If anybody knows that, all it shows is that they know who I am. If any entity assumes that knowledge of that information implies the person who knows it is me, that entity obviously has trouble getting two neurons firing in sync.
Also in this case probably a violation of the Video Privacy Protection Act. YouTube sounds like they qualify as a "video tape service provider" in (a)(4) ("delivery", "similar audio visual materials"). (b)(2)(F) covers disclosure in a civil proceeding, Viacom hasn't given prior notice to consumers of their request for disclosure and YouTube users haven't been given the opportunity to appear to contest Viacom's request, both are required by that paragraph (the following part about appropriate safeguards applies only if the order's granted pursuant to (F), and since the requirements of (F) weren't met the order couldn't have been properly issued pursuant to it).
There's actually a solution to that, but it involves slowing the process down. Just don't activate the account once the information's entered. Instead, send a physical letter to the credit-card billing address. You can require a form to be signed and returned, or just include an activation code in the letter that has to be entered to turn the account on. That should make it infeasible to use 99% of stolen cards. It introduces a few days of delay between requesting the account and getting it, but IMO if you intend to use the account for any length of time a few days shouldn't be an issue and if you don't then you're likely exactly the kind of person this is intended to filter out.
Amazon has billing information for those accounts. Money changes hands. So, require that the name and address given to Amazon when setting up the account match the billing name and address of the credit card used to pay for the services. Most mail-order and on-line merchants do that already, and won't ship except to the billing name/address. Then block known pre-paid debit card numbers and one-time card numbers. Not perfect, but it should knock down 90% of the problems and make it a lot harder for a criminal to get a new account not tied to a consistent identity.
If I wanted to keep my code private, then I wouldn't use GPL'd code as the basis for my firmware, simple as that. When I use someone else's code, I have to abide by their license terms. If I can't or won't do so, I don't use their code. Nothing specific to GPL'd software there, the same thing applies to commercial software. If I don't want to or can't pay the royalties MS requires for Windows CE, I don't use Windows CE in my devices.
If your only complain is "I can't use someone else's software without complying with their terms!", then I think you'll get precious little sympathy here.
I think the key question would be this: can you modify the code on the board without a dev kit? If you've got some way of getting your code onto the board without a dev kit, or by using some piece of hardware or software the user's prohibited from obtaining, then you may run afoul of the GPLv3's terms. But if all you need's a dev kit, both you and the user would need that dev kit to change the software and you aren't prohibiting the user from purchasing that dev kit, then you'll be OK. GPLv3 doesn't require that you let the user modify the code running on the device, just that you not block them from access while retaining that access yourself. As someone else noted, if you burned the code into ROM then the user can't modify it, but that's acceptable under the GPLv3 because you can't modify it either (short of burning a new ROM, going out and physically installing it, which the user could do too if they were so inclined and capable).
Just bear in mind that this'll prevent proper handling of hosts that use tcp_wrappers and the inetd superserver to handle services and access control. That combination works by accepting the connection, getting the peer address (must be connected for this), checking the access rules and immediately issuing an RST before any data's been transferred if access is denied. This rule prevents that RST from getting through, leaving you handing trying to connect to a port that's not going to allow you to connect.
And so it comes full circle. The ancient flat hosts file that the ARPAnet used way back in it's early days, the one that was abandoned in favor of hierarchical DNS because it wasn't possible to manage a flat namespace when the net was a few thousand machines, returns. What, you think the companies that insist on registering every variation of their name in every domain in existence won't insist on having their own TLD too? And we'll be back to hostnames being of the form "ford". And "fordfocus". And "focus". And "myfocus". And "myford". All belonging to Ford Motors. And the inevitable fights when Focus Magazine (a fine-art photography magazine) also wants "focus" and has the trademark to justify getting it just as much as Ford.
Siiiiiigh...
And as any lawyer, or any businessman, will tell you, when the other party says "There's no problem, we don't need to put that in writing.", you will be needing it in writing at some point and you'll be regretting not getting it.
LimitNone's only hope is language in a written contract promising that Google won't compete with their product. Absent that, they're toast. Let's face it, their product isn't much. There's nothing in it that hasn't been well-known and in wide use for the last 30 years, and thus can't be a trade secret. Google obviously knows their own formats and APIs for loading messages into gMail. In fact they had to have created them before LimitNone's product existed, if they hadn't LN couldn't have created their product at all. The file formats and things like Exchange server APIs aren't exactly secret. Copying all messages from one to the other is the same basic copy loop that's been used for 40-some years: open input; while not eof(input) do read(input); write(output); done; close input. A loop to iterate through folders and some recursion to handle subfolders, I was doing that in high school. Look and feel? LimitNone's probably using the standard tree-view widgets provided by the system, so yes Google's app will look like theirs because both of them look like the standard system widgets. That's assuming the apps allow message-level selectivity, if they limit it to folder-level or "everything on this server", the UI's even more generic. And the concept of importing mail messages from an old client into a new one? Hardly new. Mail clients have been importing other client's mailboxes since as long as there've been mail clients. Thunderbird has been doing it IIRC for a couple years now, well before LimitNone's product was created.
LimitNone's problem is that they're trying to charge $29 for a basic one-shot function that comes standard with most mail clients and that frankly could be hacked up by a single programmer in a few weeks of full-time work.
Yes, but that's the point: everybody does know (or any reasonable person should know) that that happens. Servers cease to exist, servers move to other names/locations, the people running the server decide they don't want to host older versions anymore when newer ones are available. And while the courts will tend to do what you suggest when someone couldn't reasonably have known, they're less tolerant of people who knew full well what would happen and did it anyway.
As far as being nice to companies, I ask this: if you broke their license knowing full well what you were doing, how would they treat you? And why should they expect to be treated any differently than they'd treat someone in their position? That's my touchstone: I'll cut them exactly as much slack as they'd cut anyone else. Which is usually precious little. And if I were running a business, I'd simply avoid the whole problem by either complying with the license or not distributing that software. Then the problem never comes up and I don't have to worry about whether anybody'll cut me any slack.
Well, the BusyBox people aren't bound by the GPL's terms. They're the copyright owners, they don't need a license to distribute their own works. They could stop tomorrow, take everything down, close up shop without any further legal obligations to anyone. Only the people who redistributed their code are under any legal obligation to continue providing the source code.
And GPL v2 is even worse. Under it offering access from a network server is not sufficient to satisfy it's section 3b requirements. If you distribute via section 3a and distribute only by offering download from a server then you can offer the sources from the same server and location as the binaries and be OK, but 3b doesn't allow that option in GPL v2. If someone asks, you have to provide the source on a distribution medium. Telling them to go to someone else isn't providing the source code to them on a distribution medium, I'm afraid.
Except for one small detail: the GPL doesn't require that the source code be available. It requires that the party doing the distributing (or redistributing) make the source code available. Read section 3 of the GPL v2 or section 6 of the GPL v3. The clearest statement is in GPL v3 section 6d, where it says clearly "Regardless of what server hosts the corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.". The servers hosting the BusyBox code aren't obliged to keep version 1.n.n up, and if you depend on them and they take it down you are the one in violation of the GPL by not being able to provide access to the source code as required.
You're correct in thinking that. You don't have to provide access to just any source code for what you distributed, you have to provide source code for the version you distributed. If you don't have an agreement with the party hosting the source code you point to, they may update to a newer version or cease hosting it completely without any warning to you. And since you are the one obligated to provide the source code or access to it, not them, you'll be the one on the hook for copyright violation if/when that happens. And I'd note that it likely wouldn't be considered "unwitting" if the case went to court, since any reasonable person would know or should have known that this would occur.
IANAL, but every time I've talked to one "distribution" depends on who it's being distributed to. If I'm at both ends of the transaction, it is not distribution in the sense the GPL uses, any more than it's distribution in the GPL sense if I copy the code from my desktop box to the server box sitting next to it and plugged into the same Ethernet switch just above them. Physical distance doesn't matter, it's ownership/control of the ends and if I own or have sole control over both ends then it's just me and there's no distribution involved in the legal sense. Legal here is just being paranoid, or they're somewhat clueless about networks and software (not a good thing for the lawyers of a company that deals with both, IMO).
Now, if the servers aren't under your sole control, if it's a shared system where other entities than your company have root access and/or significant control over the systems without your supervision, that would be a different case.
Except that some people don't realize what's happening. They put a file into their shared folder so they can copy it to another of their computers on the network, not realizing that in the process it's being made visible to everyone else who looks.
Analogous situation: I like to read out on the porch, so I've got a stack of books on the table on the porch so they're conveniently reachable. I hadn't realized that someone else had left the gate open so anybody wandering by could walk in and make a copy of any book they wanted and walk away with it. Certainly there'd be no point in my making the books available to everybody else, except that I didn't realize I was making them available. And gods know the majority of people are computer-clueless and have no idea what a "shared folder" is, they just think "No, that isn't a shared folder, that's just the folder I put things in when I need to copy them to another computer. I'm not sharing with anybody, it's all my computers.".
Possibly you're right. However, I'm minded of something that happened when I was living in rural Nevada. One of the (newish) salesmen for the Ford dealer saw a cowboy wandering around looking at the luxury cars on the lot. The cowboy was a real rough-looking type, dirty, sweaty, holes in his jeans, looked about 60 and was driving a half-rusted-out pickup. The salesman ignored him, and after a while the cowboy wandered over and asked about one of the more expensive cars. The salesman blew him off, figured this guy wasn't going to be able to afford that car, not as poor as he looked. The cowboy left.
The next day, the dealership owner calls everybody in. He hauls the salesman above up in front of everybody and chews him out royally, ending with "You're fired! Get off my lot, NOW!". It turns out that one of the biggest ranchers in the area, a guy who doesn't bother writing checks for less than 10 grand because he's got that much in cash in his pocket on any given day, who'd been buying all his ranch's vehicles from this dealership for years, just called in to say he was ending all his leases and taking his business down the road to the Dodge dealer. That added up to a lot of sales lost. Apparently the rancher had been in the day before looking for a new car, with cash in his pocket ready to buy on the spot, and the salesman wouldn't even give him the time of day.
When you're deciding that it's not worth your time to pay attention to something, be very very sure you've taken everything into account.
Already covered. GPL v2 section 7:
f, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.
If you can't for some reason make available all the tools needed to build your derivative work, and those tools are not standard parts of a system or otherwise available independently, then you won't be able to comply with the GPL and thus won't be able to distribute while you're required to comply with the GPL. However, if you were using only the standard Microsoft compilers and such, they'd probably fall into the "standard system software" category just like GCC does (one must distribute the makefiles needed to build a GPL'd work, but one doesn't need to distribute GCC and make along with it).
Because it isn't their application. It's the source to the BusyBox author's application, which they're distributing without permission to do so (they aren't complying with the terms of the only license which would give them permission). I think it ought to come as no surprise to any sensible adult that you don't get to do anything you like with someone else's stuff.
Odds on the developers understand the GPL and it's obligations completely. But the packaging and distribution of the product isn't handled by the developers, it's handled by Marketing and Sales. And those guys probably don't even know the software has a license attached, they've got no clue what all went into their software, and they likely don't think they ought to consult with mere software developers about how they can market the product. And they probably didn't ask Legal for an opinion, since it's "their" software and they can (in Marketing's world) do whatever they want with it. Customer Support's likely a division of Marketing, so when the initial e-mails came in they got handled by people with that attitude.
And then one day the package with the letter on a lawyer's letterhead with all the court paperwork arrived, and it went straight to Legal, bypassing Marketing entirely. And Legal, being sensible people, asked the obvious questions of the obvious people. Marketing may ignore the engineers, but Management tends to listen to the lawyers.
Almost no one out side the free software community uses that definition of "free"? Oh, so when the star of The Prisoner said "I am not a number! I am a free man!", he meant he didn't cost any money? Oh, oops, no he didn't. The definition the free software community uses is, I'm afraid, one of the two common definitions for the word "free" that are in common use today. That you'd rather it only had one meaning, and that the monetary one, doesn't change the dictionary.
GPL v2 section 3b says otherwise:
Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange;
See the bolded words? They don't say "any direct recipient" or "anyone who got the binaries from you", they say "any third party". Remember that the people who received binaries from you may have redistributed them under section 3c, in which case those people who didn't get the binaries directly from you are still entitled under the GPL to receive source code from you (by taking advantage of the written offer the original recipient passed along to them).
We've covered this time and time again. Why do people keep insisting that the GPL doesn't say what it says?