The example is meant to be a case of mere aggregation, as described in the GPL. The example is so short that it may not get the point across. Here is some more information.
http://www.gnu.org/licenses/gpl-faq.html#MereAggre gation says:
---
"Mere aggregation of two programs means putting them side by side on the same CD-ROM or hard disk. We use this term in the case where they are separate programs, not parts of a single program. In this case, if one of the programs is covered by the GPL, it has no effect on the other program.
Combining two modules means connecting them together so that they form a single larger program.... What constitutes combining two parts into one program?... We believe that a proper criterion depends both on the mechanism of communication (exec, pipes, rpc, function calls within a shared address space, etc.) and the semantics of the communication (what kinds of information are interchanged).... pipes, sockets and command-line arguments are communication mechanisms normally used between two separate programs. So when they are used for communication, the modules normally are separate programs. But if the semantics of the communication are intimate enough, exchanging complex internal data structures, that too could be a basis to consider the two parts as combined into a larger program."
---
The given examples are probably meant to be separate programs invoking each other by exec, and with no complex internal data structures passed, so a distribution that contained both would be a mere aggregation.
There is a continuum of possibilities for mere aggregation versus "parts of a single program". For cases close to the border, which side applies would have to be decided by negotiation or by a court.
Distributing a collection containing GPL'd works
on
Open Source Licensing
·
· Score: 2, Interesting
If someone distributes a stand-alone proprietary program that is designed to dynamically link to a GPL'd program, and does not distribute the GPL'd program, then, I think, there is no problem with the GPL.
However, if this same program is distributed in a collection containing the GPL'd program, then there is a problem: the GPL does not give permission to distribute the collection, since it is not a mere aggregation and contains a non-GPL'd program and a GPL'd program.
SCO sues Novell, but won't disclose its copyrights in Linux, even though it has to defy a court order.
In its press releases and letters to its customers, SCO claims that violations of its copyrights are in the Linux kernel. On December 5,SCO's lawyer, Keven McBride, said in court that SCO would be filing copyright claims against IBM. He also objected to the requirement to specify copyrights, but, after the judge insisted, said that SCO would comply. SCO was compelled by court order to answer ALL IBM questions with specificity by January 12. One of those questions was:
INTERROGATORY NO. 12: Please identify, with specificity (by file and line of code), (a) all source code and other material in Linux (including but not limited to the Linux kernel, any Linux operating system and any Linux distribution) to which plaintiff has rights; and (b) the nature of plaintiff's rights, including but not limited to whether and how the code or other material derives from UNIX.
On 1/12/04, SCO responded with an affidavit, a 60 page supplemental response, and some documents. Like everyone else, except IBM, I am still waiting to see the 60 page supplemental response, but I have additional information from public statements by SCO.
On 1/13/04, in the interview "SCO shows IBM the code", SCO spokesman Bruce Stowell said: "Monday's response included no examples of copyright violations. We've not introduced copyright infringement as part of our case with IBM. We've tried to make it clear that it's a contract issue."
SCO could have made arguments like this before it was ordered to respond to IBM's questions without further discussion. Now it's too late.
The order said to identify all rights that SCO claimed. It was not limited to rights that SCO was currently claiming as part of its suit against IBM, and, in my opinion, even includes the code contributed by SCO to Linux.
If, as Stowell said, SCO did not specify any copyrights that it owns in Linux, then either it is not claiming to own any such copyrights, or it has not complied with the court order.
If SCO now officially declines to claim ownership of any such copyrights, then it is hard to see how they could claim them in any later suit against a Linux user or distributor.
If SCO does claim ownership of any copyrights in Linux, it has blatantly defied the court order, and the judge can order sanctions. My guess is that part of the sanctions would be to bar SCO from suing anyone on any claims that it failed to specify in its 1/12 response to the compel order. The judge might also order SCO to show cause why it should not be held in contempt, and why it should not suffer sanctions for failing to comply with the order.
I wonder if SCO will now claim that it does not own any code in Linux, but it owns the copyrights that Novell claims. It would be typical of the way SCO has been making and revising claims.
The Protective Order was agreed to long ago by IBM and SCO, and ratified by the court. It provides that either party can label a discovery item confidential, but the other party can challenge, and, if they cannot agree, the judge will decide whether the item requires secrecy. Until the judge decides, the item has to be treated as confidential. That is why the hearing has to be closed until the judge rules on this question.
IBM will certainly challenge the items that don't really require secrecy, and the Protective Order provides that, in that case, SCO has the burden of proving that there is a good reason for keeping the items secret, one by one.
I don't think that the judge issued an independent order saying that the hearing will be closed; no such order appeared on the court docket. I am sure that the judge has not ruled that any of the discovery items are confidential; the arguments have not been presented yet.
Stowell lied about the import of the closed hearing, and lied again when he implied that all the results of discovery would be secret until 2005 and beyond. IMO, the most that will be left secret after the hearing will be parts of SCO's code that SCO does not claim were copied into Linux.
Further, if any trade secrets were illegally copied by IBM into Linux, they are no longer secret. IBM might have to pay damages to SCO, but the information is no longer secret, and SCO cannot collect from anyone else. Again, the contract I looked at said that IBM could refer to SCO's code and use SCO's ideas and methods while developing IBM's products, and IBM would own those products. I saw nothing that implied that IBM had to keep its own products secret.
Lawyers: up to $1 million + 20%. Clients: jail
on
SCO's Lawyers Analyzed
·
· Score: 2, Interesting
SCO's lawyers get up to $1,000,000 plus 20% of essentially everything else SCO owns: stock, sale, settlement, winnings in court, license fees. The clients get a bankrupt company and jail terms.
SCOX is currently worth $17 per share, so the lawyer would get $49 million, if they could sell it quickly enough.
No one would buy the company, except as a gift to SCO (Microsoft?), and then they would get a can of worms with a 20% surcharge.
IBM won't settle, and won't buy.
Winnings in court? Don't make me laugh.
License fees: SCO has already collected almost all the license fees it is going to get.
How quickly would the lawyers have to collect in order for the 20% to exceed the up front payment? Well, let's consider some recent history and scheduled events.
SCO's press releases have become few and far between; each is quickly answered by someone from SlashDot or Groklaw, sometimes in a Talkback forum on the same medium.
IBM's motion to compel discovery will be heard in court on December 5, and I expect the judge to rule in favor of IBM.
SCO's motion to dismiss the Redhat suit will come up soon, and I expect the judge to rule in favor of Redhat.
Novell bought SuSe, greatly enhancing its multi-platform services strategy, and making it a major Linux player, both for services and for desktop. The Linux community now has two very strong competitors, Redhat and Novell, with giant IBM backing both.
Novell and its customers are effectively immune from suit by SCO, because of the rights Novell retained when it sold assets to SCO. NOTE: No indemnification or license from SCO is needed by any Novell customer.
IBM invested exactly $50 million in Novell, matching Microsoft's gift to SCO. This has to be a hint to Novell to file a suit or criminal complaint against SCO.
I filed a criminal complaint with the SEC, against SCO's officers and directors, in September, accusing them of stock market manipulation. HAVE YOU?
How quickly would the lawyers have to collect in order for the 20% to exceed the up front payment? My guess is six weeks maximum.
I don't know when the criminal complaints will become public knowledge, but I expect that to be the last straw for SCO.
Everyone should be able to watermark software. Further, forging a watermark should be a felony.
This is not a matter of good guys vs bad guys, or open source vs closed source. It is a matter of protecting everyone's intellectual property from thieves, GPL'd code, and Microsoft's code alike.
I like RH9's approach to integrating Gnome and KDE, but it has much further to go. Some like KDE better, but I started with Gnome, and have not gotten to try KDE yet. It would help if RH integrated them further, say by a better organization of the Start menu, with the main entries listing a tree of groups of tasks meaningful to the user, each leaf pointing to RH's opinion of the best application for a task, and side entries listing a tree of alternative applications.
For related discussion, see the following threads on comp.os.advocacy. A few of my quotes and one other are listed.
A. Windows "Easier". "99.999% of user's wouldn't know where to start [to fix a boot sector problem]".
B. A case for a standard Linux GUI. "What good wi[ll] 10 or perhaps 100 million linux users do when there are billions (potentially), who will still shell out do$h for M$ software?"
C. How can Linux even be considered viable for the desktop user? "But, when he or she is considering buying a new house (Windows 2003) from someone who wants to collect monthly rental after the purchase, he or she might want to consider a cheap addition (dual boot Linux), which might turn out to be so comfortable (stability, performance, security, scalability, etc.) that he or she might decide to live full time in the addition (uninstall Windows)."
D. Linux ready for the desktop. "I agree that Linux is not yet ready for the desktop, even though I have been using Linux as my desktop system for nearly a year. I feel that I am expert in some features of Linux, a novice in most.
For example, even though I have contributed quite a bit of code to the gnucash port to Gnome 2.2, I have yet to run the Gnome 1 version, and have
not gotten beyond the first few screens in the Gnome2 version. Next year I hope that gnucash will be running on Gnome 2.2, or better, and I expect to be an expert at using it as well as developing it. I hope to use gnucash as an example of how far Linux has come in getting ready for the desktop."
"the only voice most businesses (and individuals) hear is SCO's"
Note that SCO has gotten very quiet the last few days, and the open source community has gotten very loud. See the following web page.
http://www.opensource.org/halloween/halloween9.htm l . (please check that "html" is not split).
Otherwise, I agree. We should refute their lies and condemn their crimes in no uncertain terms.
Slashdot Mirror
http://www.gnu.org/licenses/gpl-faq.html#MereAggre gation says:
---
"Mere aggregation of two programs means putting them side by side on the same CD-ROM or hard disk. We use this term in the case where they are separate programs, not parts of a single program. In this case, if one of the programs is covered by the GPL, it has no effect on the other program.
Combining two modules means connecting them together so that they form a single larger program. ... What constitutes combining two parts into one program? ... We believe that a proper criterion depends both on the mechanism of communication (exec, pipes, rpc, function calls within a shared address space, etc.) and the semantics of the communication (what kinds of information are interchanged).... pipes, sockets and command-line arguments are communication mechanisms normally used between two separate programs. So when they are used for communication, the modules normally are separate programs. But if the semantics of the communication are intimate enough, exchanging complex internal data structures, that too could be a basis to consider the two parts as combined into a larger program."
---
The given examples are probably meant to be separate programs invoking each other by exec, and with no complex internal data structures passed, so a distribution that contained both would be a mere aggregation.
There is a continuum of possibilities for mere aggregation versus "parts of a single program". For cases close to the border, which side applies would have to be decided by negotiation or by a court.
If someone distributes a stand-alone proprietary program that is designed to dynamically link to a GPL'd program, and does not distribute the GPL'd program, then, I think, there is no problem with the GPL.
However, if this same program is distributed in a collection containing the GPL'd program, then there is a problem: the GPL does not give permission to distribute the collection, since it is not a mere aggregation and contains a non-GPL'd program and a GPL'd program.
SCO sues Novell, but won't disclose its copyrights in Linux, even though it has to defy a court order.
In its press releases and letters to its customers, SCO claims that violations of its copyrights are in the Linux kernel. On December 5,SCO's lawyer, Keven McBride, said in court that SCO would be filing copyright claims against IBM. He also objected to the requirement to specify copyrights, but, after the judge insisted, said that SCO would comply. SCO was compelled by court order to answer ALL IBM questions with specificity by January 12. One of those questions was:
INTERROGATORY NO. 12: Please identify, with specificity (by file and line of code), (a) all source code and other material in Linux (including but not limited to the Linux kernel, any Linux operating system and any Linux distribution) to which plaintiff has rights; and (b) the nature of plaintiff's rights, including but not limited to whether and how the code or other material derives from UNIX.
On 1/12/04, SCO responded with an affidavit, a 60 page supplemental response, and some documents. Like everyone else, except IBM, I am still waiting to see the 60 page supplemental response, but I have additional information from public statements by SCO.
On 1/13/04, in the interview "SCO shows IBM the code", SCO spokesman Bruce Stowell said: "Monday's response included no examples of copyright violations. We've not introduced copyright infringement as part of our case with IBM. We've tried to make it clear that it's a contract issue."
SCO could have made arguments like this before it was ordered to respond to IBM's questions without further discussion. Now it's too late.
The order said to identify all rights that SCO claimed. It was not limited to rights that SCO was currently claiming as part of its suit against IBM, and, in my opinion, even includes the code contributed by SCO to Linux.
If, as Stowell said, SCO did not specify any copyrights that it owns in Linux, then either it is not claiming to own any such copyrights, or it has not complied with the court order.
If SCO now officially declines to claim ownership of any such copyrights, then it is hard to see how they could claim them in any later suit against a Linux user or distributor.
If SCO does claim ownership of any copyrights in Linux, it has blatantly defied the court order, and the judge can order sanctions. My guess is that part of the sanctions would be to bar SCO from suing anyone on any claims that it failed to specify in its 1/12 response to the compel order. The judge might also order SCO to show cause why it should not be held in contempt, and why it should not suffer sanctions for failing to comply with the order.
I wonder if SCO will now claim that it does not own any code in Linux, but it owns the copyrights that Novell claims. It would be typical of the way SCO has been making and revising claims.
The Protective Order was agreed to long ago by IBM and SCO, and ratified by the court. It provides that either party can label a discovery item confidential, but the other party can challenge, and, if they cannot agree, the judge will decide whether the item requires secrecy. Until the judge decides, the item has to be treated as confidential. That is why the hearing has to be closed until the judge rules on this question.
IBM will certainly challenge the items that don't really require secrecy, and the Protective Order provides that, in that case, SCO has the burden of proving that there is a good reason for keeping the items secret, one by one.
I don't think that the judge issued an independent order saying that the hearing will be closed; no such order appeared on the court docket. I am sure that the judge has not ruled that any of the discovery items are confidential; the arguments have not been presented yet.
Stowell lied about the import of the closed hearing, and lied again when he implied that all the results of discovery would be secret until 2005 and beyond. IMO, the most that will be left secret after the hearing will be parts of SCO's code that SCO does not claim were copied into Linux.
Further, if any trade secrets were illegally copied by IBM into Linux, they are no longer secret. IBM might have to pay damages to SCO, but the information is no longer secret, and SCO cannot collect from anyone else. Again, the contract I looked at said that IBM could refer to SCO's code and use SCO's ideas and methods while developing IBM's products, and IBM would own those products. I saw nothing that implied that IBM had to keep its own products secret.
SCO's lawyers get up to $1,000,000 plus 20% of essentially everything else SCO owns: stock, sale, settlement, winnings in court, license fees. The clients get a bankrupt company and jail terms. SCOX is currently worth $17 per share, so the lawyer would get $49 million, if they could sell it quickly enough. No one would buy the company, except as a gift to SCO (Microsoft?), and then they would get a can of worms with a 20% surcharge. IBM won't settle, and won't buy. Winnings in court? Don't make me laugh. License fees: SCO has already collected almost all the license fees it is going to get. How quickly would the lawyers have to collect in order for the 20% to exceed the up front payment? Well, let's consider some recent history and scheduled events. SCO's press releases have become few and far between; each is quickly answered by someone from SlashDot or Groklaw, sometimes in a Talkback forum on the same medium. IBM's motion to compel discovery will be heard in court on December 5, and I expect the judge to rule in favor of IBM. SCO's motion to dismiss the Redhat suit will come up soon, and I expect the judge to rule in favor of Redhat. Novell bought SuSe, greatly enhancing its multi-platform services strategy, and making it a major Linux player, both for services and for desktop. The Linux community now has two very strong competitors, Redhat and Novell, with giant IBM backing both. Novell and its customers are effectively immune from suit by SCO, because of the rights Novell retained when it sold assets to SCO. NOTE: No indemnification or license from SCO is needed by any Novell customer. IBM invested exactly $50 million in Novell, matching Microsoft's gift to SCO. This has to be a hint to Novell to file a suit or criminal complaint against SCO. I filed a criminal complaint with the SEC, against SCO's officers and directors, in September, accusing them of stock market manipulation. HAVE YOU? How quickly would the lawyers have to collect in order for the 20% to exceed the up front payment? My guess is six weeks maximum. I don't know when the criminal complaints will become public knowledge, but I expect that to be the last straw for SCO.
Everyone should be able to watermark software. Further, forging a watermark should be a felony. This is not a matter of good guys vs bad guys, or open source vs closed source. It is a matter of protecting everyone's intellectual property from thieves, GPL'd code, and Microsoft's code alike.
I like RH9's approach to integrating Gnome and KDE, but it has much further to go. Some like KDE better, but I started with Gnome, and have not gotten to try KDE yet. It would help if RH integrated them further, say by a better organization of the Start menu, with the main entries listing a tree of groups of tasks meaningful to the user, each leaf pointing to RH's opinion of the best application for a task, and side entries listing a tree of alternative applications. For related discussion, see the following threads on comp.os.advocacy. A few of my quotes and one other are listed. A. Windows "Easier". "99.999% of user's wouldn't know where to start [to fix a boot sector problem]". B. A case for a standard Linux GUI. "What good wi[ll] 10 or perhaps 100 million linux users do when there are billions (potentially), who will still shell out do$h for M$ software?" C. How can Linux even be considered viable for the desktop user? "But, when he or she is considering buying a new house (Windows 2003) from someone who wants to collect monthly rental after the purchase, he or she might want to consider a cheap addition (dual boot Linux), which might turn out to be so comfortable (stability, performance, security, scalability, etc.) that he or she might decide to live full time in the addition (uninstall Windows)." D. Linux ready for the desktop. "I agree that Linux is not yet ready for the desktop, even though I have been using Linux as my desktop system for nearly a year. I feel that I am expert in some features of Linux, a novice in most. For example, even though I have contributed quite a bit of code to the gnucash port to Gnome 2.2, I have yet to run the Gnome 1 version, and have not gotten beyond the first few screens in the Gnome2 version. Next year I hope that gnucash will be running on Gnome 2.2, or better, and I expect to be an expert at using it as well as developing it. I hope to use gnucash as an example of how far Linux has come in getting ready for the desktop."
"the only voice most businesses (and individuals) hear is SCO's" Note that SCO has gotten very quiet the last few days, and the open source community has gotten very loud. See the following web page. http://www.opensource.org/halloween/halloween9.htm l . (please check that "html" is not split).
Otherwise, I agree. We should refute their lies and condemn their crimes in no uncertain terms.