Loopholes

You’re of course correct in your conclusion that “the GPL specifically and deliberately permits distributors to ‘ship proprietary binaries on the same CD as free software.’” There is no ambiguity on this point. Distributions have been shipping proprietary binaries on the same CD as free software almost since Linux distributions were invented (early 90s), and the v3 makes no attempt to stop that, nor has the FSF ever attempted to stop it by other means.

Given that this history is now creeping up on 20 years, I think maybe you forgot option (c), ‘Blafkin is deliberately misinterpreting the position taken by FSF over a nearly 20 year period in order to score cheap political points.’

You’re of course correct in your conclusion that “the GPL specifically and deliberately permits distributors to ‘ship proprietary binaries on the same CD as free software.’” There is no ambiguity on this point. Distributions have been shipping proprietary binaries on the same CD as free software almost since Linux distributions were invented (early 90s), and the v3 makes no attempt to stop that, nor has the FSF ever attempted to stop it by other means.

Given that this history is now creeping up on 20 years, I think maybe you forgot option (c), ‘Blafkin is deliberately misinterpreting the position taken by FSF over a nearly 20 year period in order to score cheap political points.’

Oh Tim… there you go again! You’re awfully punchy this afternoon.

You got me. I may have misused the term “loophole.” I bow before your superior use of Dictionary.com.

Let me try to explain it differently. The GPL is written to protect Stallman’s Four Freedoms. Section 2b of the GPL is specifically designed to prevent distribution of GPL programs that integrate non-free code into GPL code, because this would inherently limit the users freedoms. Section 2b says that you can copy and distribute GPL programs provided that:

You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

Think about it this way. This provision essentially says that you can’t distribute cakes that have proprietary milk and GPL eggs in them. It doesn’t, however, prevent the distribution of a box that includes that proprietary milk, GPL eggs, BSD flour, Apache sugar, and specific directions on how to turn them into a cake. That ability is protected by the provision that you cite.

However, does that practice (i.e. The shipping of separate proprietary video drivers with Ubuntu Linux with the intent of mixing them together on the end machine) really fit within the spirit of the GPL and its Four Freedoms?

No Way! How do I know that? Because Richard Stallman and Eben Moglen have said it over and over again. Here are just a few examples:

In a 2006 CNET article, Eben Moglen said very specifically that:

If the kernel’s terms were unambiguously GPL, which they are apparently not, (proprietary drivers) would be forbidden.

In addition, Stallman told an interviewer that:

To “argue” in favor of adding non-free software in GNU/Linux distros is almost superfluous, since that’s what nearly all of them have already done. This reflects the general spinelessness of our community. Most of its members have never heard the philosophy of freedom and community which motivated the GNU Project to build the community, and most care more about convenience than freedom.

In his piece on Free Software after 20 years, Stallman wrote that:

Today we have a large community of users who run GNU, Linux and other free software. Thousands of people would like to extend this, and have adopted the goal of convincing more computer users to “use free software”. But what does it mean to “use free software”? Does that mean escaping from proprietary software, or merely installing free programs alongside it? Are we aiming to lead people to freedom, or just introduce them to our work? In other words, are we working for freedom, or have we replaced that goal with the shallow goal of popularity?

The GPL explicitly allows the distributors to ship separate binaries on the same CD that include GPL and proprietary software. The fact that some distributors use that to create a cake for the end user is clearly something Stallman did NOT intend. Call it what you will, but it still sounds like a loophole to me.

So to summarize, Mark: Richard Stallman doesn’t like proprietary software, he hopes to persuade consumers to stop using it, but he hasn’t tried to use the GPL to force them to stop doing so. Or, as I put it two weeks ago:

The FSF has never tried to prevent free software companies from cooperating with proprietary software companies to make their systems interoperable. They still prefer that the user using proprietary software switch to free software, and they might not approve of collaboration between proprietary and free software companies, but the GPL doesn’t make any attempt to prevent such cooperation.

You’ve found lots of quotes indicating that Richard Stallman doesn’t like proprietary software, which was never in doubt. What you haven’t done is explain how the GPL “actively attempts to limit collaboration between proprietary and free software communities.”

Indeed, the only kind of “collaboration” that it tries to limit is the kind where a proprietary software company helps themselves to free software code and incorporates it into their proprietary products. Microsoft certainly wouldn’t tolerate it if Linux developers “collaborated” with Microsoft by adding Windows source code into the Linux kernel without permission. Why doesn’t the same principle apply when the shoe is on the other foot?

Oh Tim… there you go again! You’re awfully punchy this afternoon.

You got me. I may have misused the term “loophole.” I bow before your superior use of Dictionary.com.

Let me try to explain it differently. The GPL is written to protect Stallman’s Four Freedoms. Section 2b of the GPL is specifically designed to prevent distribution of GPL programs that integrate non-free code into GPL code, because this would inherently limit the users freedoms.
Section 2b says that you can copy and distribute GPL programs provided that:

You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

Think about it this way. This provision essentially says that you can’t distribute cakes that have proprietary milk and GPL eggs in them. It doesn’t, however, prevent the distribution of a box that includes that proprietary milk, GPL eggs, BSD flour, Apache sugar, and specific directions on how to turn them into a cake. That ability is protected by the provision that you cite.

However, does that practice (i.e. The shipping of separate proprietary video drivers with Ubuntu Linux with the intent of mixing them together on the end machine) really fit within the spirit of the GPL and its Four Freedoms?

No Way! How do I know that? Because Richard Stallman and Eben Moglen have said it over and over again. Here are just a few examples:

In a 2006 CNET article, Eben Moglen said very specifically that:

If the kernel’s terms were unambiguously GPL, which they are apparently not, (proprietary drivers) would be forbidden.

In addition, Stallman told an interviewer that:

To “argue” in favor of adding non-free software in GNU/Linux distros is almost superfluous, since that’s what nearly all of them have already done. This reflects the general spinelessness of our community. Most of its members have never heard the philosophy of
freedom and community which motivated the GNU Project to build the community, and most care more about convenience than freedom.

In his piece on Free Software after 20 years, Stallman wrote that:

Today we have a large community of users who run GNU, Linux and other free software. Thousands of people would like to extend this, and have adopted the goal of convincing more computer users to “use free software”. But what does it mean to “use free software”? Does that mean escaping from proprietary software, or merely installing free programs alongside it? Are we aiming to lead people to freedom, or just introduce them to our work? In other words, are we working for freedom, or have we replaced that goal with the shallow goal of popularity?

The GPL explicitly allows the distributors to ship separate binaries on the same CD that include GPL and proprietary software. The fact that some distributors use that to create a cake for the end user is clearly something Stallman did NOT intend. Call it what you will, but it still sounds like a loophole to me.

Wow, you remind me of my ex-girlfriend. No matter what I say, you only hear what you want to hear, and refuse to ever admit when you’re wrong.

What part of Eben Moglen’s legal opinion that, “If the kernel’s terms were unambiguously GPL, which they are apparently not, (proprietary drivers) would be forbidden” do you not understand?

Let me try to explain it like a computer program:

IF Program = 100% GPL

THEN No Proprietary Code Allowed

GOTO IF

REPEAT

Please excuse my syntax, it’s been awhile.

Perhaps the problem is that you don’t really understand the word “collaboration.” As defined by Dictionary.com:

To work together, especially in a joint intellectual effort.

This is about working together on a single effort like a novel, not slapping together a couple of random short stories into collection. Interoperability is a completely different concept too, so please stop trying to conflate them.

Please try re-reading my reply, especially the cake analogy again, because I think is really instructive. If you take the time to try to understand, you’re surely realize that the GPL not only prevents Microsoft from expropriating GPL code (which is great and it is clearly one goal of the license), but it ALSO makes it very difficult for collaborative projects that include both GPL and non-GPL code. Perhaps you should take this conversation up with Eben Moglen, I’m sure he would set you straight.

So to summarize, Mark: Richard Stallman doesn’t like proprietary software, he hopes to persuade consumers to stop using it, but he hasn’t tried to use the GPL to force them to stop doing so. Or, as I put it two weeks ago:

The FSF has never tried to prevent free software companies from cooperating with proprietary software companies to make their systems interoperable. They still prefer that the user using proprietary software switch to free software, and they might not approve of collaboration between proprietary and free software companies, but the GPL doesn’t make any attempt to prevent such cooperation.

You’ve found lots of quotes indicating that Richard Stallman doesn’t like proprietary software, which was never in doubt. What you haven’t done is explain how the GPL “actively attempts to limit collaboration between proprietary and free software communities.”

Indeed, the only kind of “collaboration” that it tries to limit is the kind where a proprietary software company helps themselves to free software code and incorporates it into their proprietary products. Microsoft certainly wouldn’t tolerate it if Linux developers “collaborated” with Microsoft by adding Windows source code into the Linux kernel without permission. Why doesn’t the same principle apply when the shoe is on the other foot?

Wow, you remind me of my ex-girlfriend. No matter what I say, you only hear what you want to hear, and refuse to ever admit when you’re wrong.

What part of Eben Moglen’s legal opinion that, “If the kernel’s terms were unambiguously GPL, which they are apparently not, (proprietary drivers) would be forbidden” do you not understand?

Let me try to explain it like a computer program:

IF Program = 100% GPL

THEN No Proprietary Code Allowed

GOTO IF

REPEAT

Please excuse my syntax, it’s been awhile.

Perhaps the problem is that you don’t really understand the word “collaboration.” As defined by Dictionary.com:

To work together, especially in a joint intellectual effort.

This is about working together on a single effort like a novel, not slapping together a couple of random short stories into collection. Interoperability is a completely different concept too, so please stop trying to conflate them.

Please try re-reading my reply, especially the cake analogy again, because I think is really instructive. If you take the time to try to understand, you’re surely realize that the GPL not only prevents Microsoft from expropriating GPL code (which is great and it is clearly one goal of the license), but it ALSO makes it very difficult for collaborative projects that include both GPL and non-GPL code. Perhaps you should take this conversation up with Eben Moglen, I’m sure he would set you straight.

Tim is right, and the quotes Mark misuses are generally taken out of context.

Yes, the FSF does promote Free Software Mark, Big Revelation there.

Yes, RMS does belief that using Free Software is a morally better choice than using non-free software.

However, as I have noted here, and elsewhere:

http://enigmafoundry.wordpress.com/2007/04/07/the-googleearth-upgrade-is-really-worth-it/

there is plenty of GPL software that can be combined with proprietary (a type of non-free software) software in a linux distribution.

Of course, Mark’s agenda is to paint the FSF as anti-freedom through a campaign of lies. Association for Competitice Technology is, in fact, a Microsoft front organization. It claims to represent many smaller companies, but they are almost all tied to Microsoft.

Look at http://www.sourcewatch.org which states:

“Association for Competitive Technology

The Association for Competitive Technology (ACT) was founded in 1998 in response to the anti-trust action against Microsoft then taking place.

It claims to lobby on behalf of “the interests of small and entrepreneurial tech companies throughout the country”, against regulation, and in favour of the controversial issue of strong Intellectual Property rights in software.

“…While ACT members include some household names like eBay, Orbitz and Microsoft, our members are primarily small and mid-size companies.” [1]

ACT has been accused of being an industry front for Microsoft, promoting a Microsoft-friendly agenda in relation to property rights and anti-trust legislation.”

Mark is basically paid to argue Microsofts talking points.

The reason I bring this up is because repeatedly fails to acknowledge Tim’s arguments, and uses misleading terms like ‘loophole’

There are two options as to why Mark uses misleading terms:

1. He is just doesn’t know the difference, or

2. He does know the difference, but because is paid by Microsoft, he is basically creating a cloud of uncertainty around points that have long been settled. The aim is not to define positions and debate honestly, but to confuse, through the sowing of fear, uncertainty and doubt over any FOSS project, such as the GPL and Linux(TM). Towards that end, Mark and ACT are just one in a long seies of pawns. SCO was another example.

Tim is right, and the quotes Mark misuses are generally taken out of context.

Yes, the FSF does promote Free Software Mark, Big Revelation there.

Yes, RMS does belief that using Free Software is a morally better choice than using non-free software.

However, as I have noted here, and elsewhere:

http://enigmafoundry.wordpress.com/2007/04/07/t…

there is plenty of GPL software that can be combined with proprietary (a type of non-free software) software in a linux distribution.

Of course, Mark’s agenda is to paint the FSF as anti-freedom through a campaign of lies. Association for Competitice Technology is, in fact, a Microsoft front organization. It claims to represent many smaller companies, but they are almost all tied to Microsoft.

Look at http://www.sourcewatch.org which states:

“Association for Competitive Technology

The Association for Competitive Technology (ACT) was founded in 1998 in response to the anti-trust action against Microsoft then taking place.

It claims to lobby on behalf of “the interests of small and entrepreneurial tech companies throughout the country”, against regulation, and in favour of the controversial issue of strong Intellectual Property rights in software.

“…While ACT members include some household names like eBay, Orbitz and Microsoft, our members are primarily small and mid-size companies.” [1]

ACT has been accused of being an industry front for Microsoft, promoting a Microsoft-friendly agenda in relation to property rights and anti-trust legislation.”

Mark is basically paid to argue Microsofts talking points.

The reason I bring this up is because repeatedly fails to acknowledge Tim’s arguments, and uses misleading terms like ‘loophole’

There are two options as to why Mark uses misleading terms:

1. He is just doesn’t know the difference, or

2. He does know the difference, but because is paid by Microsoft, he is basically creating a cloud of uncertainty around points that have long been settled. The aim is not to define positions and debate honestly, but to confuse, through the sowing of fear, uncertainty and doubt over any FOSS project, such as the GPL and Linux(TM). Towards that end, Mark and ACT are just one in a long seies of pawns. SCO was another example.

It’s pretty funny watching ACT and PFF folks try to mine Stallman and Moglen’s remarks for evidence of some sort of “legislative intent.” I think they should take a cue from Justice Scalia and not worry so much about the legislative history. Guys, if you want to understand what the GPL says, read the GPL. Interviews with the GPL’s authors are of marginal relevance at best.

I also like the way DeLong, Blafkin et. al. are almost pleading with corporate GCs to step in and draw the line against the FSF extremists and their subversive license. Apparently it’s the lawyers, not the engineers or marketeers, who are seen by PFF and ACT as the real grown-ups within the corporate world.

It’s pretty funny watching ACT and PFF folks try to mine Stallman and Moglen’s remarks for evidence of some sort of “legislative intent.” I think they should take a cue from Justice Scalia and not worry so much about the legislative history. Guys, if you want to understand what the GPL says, read the GPL. Interviews with the GPL’s authors are of marginal relevance at best.

I also like the way DeLong, Blafkin et. al. are almost pleading with corporate GCs to step in and draw the line against the FSF extremists and their subversive license. Apparently it’s the lawyers, not the engineers or marketeers, who are seen by PFF and ACT as the real grown-ups within the corporate world.

Mark, at this point I feel like we’re splitting hairs. We agree that Stallman and Moglen don’t like proprietary software. We agree that the GPL allows you to ship free software alongside proprietary software, as long as they’re not too tightly coupled together.

So we seem to be left with a semantic dispute about the meaning of the term “collaboration.” The GPL restricts those forms of collaboration that involve incorporating free software into a proprietary product. It does not restrict forms of collaboration that involve incorporating proprietary software into a free software product and releasing the whole thing under the GPL. Nor does it restrict the kind of collaboration that involves making separate free and proprietary software products work better together. In other words, the only types of “collaboration” it restricts are the ones that involve companies putting other peoples’ code into their products without permission.

There’s some disagreement within the free software movement about exactly how tightly two programs can be coupled before they become a single program for purposes of the GPL, with Moglen taking a somewhat more expansive view than Linus Torvalds, at least with respect to linking proprietary drivers into a free software kernel. But that illustrates an important fact about the GPL: the ultimate power to decide how strictly to interpret its terms is ultimately up to software developers, not the FSF. The GPL gives Linus Torvalds and his fellow kernel developers the option of restricting the use of proprietary drivers in the Linux kernel, but if they want to permit proprietary drivers, there’s nothing that Eben Moglen and Richard Stallman can do about it, because they’re not parties to the contract.

So to put it in terms of your cake analogy, if you have proprietary eggs and free flour, you can ship them together in a box with instructions on how to mix them to make a cake. Or if you own the rights to the eggs, you can combine them and release them as a free cake. Conversely, you can go to the guy (or guys) who wrote the free flour and ask them to license their code to you under a proprietary license so you can make a proprietary cake. The only thing you can’t do is put the free flour into a proprietary cake without the permission of the guy who made the flour. Or, more specifically, you couldn’t make a cake out of the flour and then get a patent on the cake to prevent anyone but you from eating it—because that effectively expropriates the “intellectual property” of the guy who made the flour.

The new version of the GPL is about closing loopholes that might allow someone to effectively put free eggs into proprietary cakes. TiVo found such a loophole when they sold devices that had embedded Linux without making the source code available. Many people believe the Microsoft-Novell agreement will wind up being another such loophole, because it could be a prelude to Microsoft suing other companies that haven’t paid Microsoft for the privilege of using Linux. These really are loopholes, because they lead to free software being transformed into unfree software: in the former case because you can’t get the source code, in the latter case because you can’t use it without paying Microsoft for the privilege. It has nothing to do with stopping collaboration, and everything to do with protecting the rights of free software developer to prevent others from expropriating their intellectual creations.

Mark, at this point I feel like we’re splitting hairs. We agree that Stallman and Moglen don’t like proprietary software. We agree that the GPL allows you to ship free software alongside proprietary software, as long as they’re not too tightly coupled together.

So we seem to be left with a semantic dispute about the meaning of the term “collaboration.” The GPL restricts those forms of collaboration that involve incorporating free software into a proprietary product. It does not restrict forms of collaboration that involve incorporating proprietary software into a free software product and releasing the whole thing under the GPL. Nor does it restrict the kind of collaboration that involves making separate free and proprietary software products work better together. In other words, the only types of “collaboration” it restricts are the ones that involve companies putting other peoples’ code into their products without permission.

There’s some disagreement within the free software movement about exactly how tightly two programs can be coupled before they become a single program for purposes of the GPL, with Moglen taking a somewhat more expansive view than Linus Torvalds, at least with respect to linking proprietary drivers into a free software kernel. But that illustrates an important fact about the GPL: the ultimate power to decide how strictly to interpret its terms is ultimately up to software developers, not the FSF. The GPL gives Linus Torvalds and his fellow kernel developers the option of restricting the use of proprietary drivers in the Linux kernel, but if they want to permit proprietary drivers, there’s nothing that Eben Moglen and Richard Stallman can do about it, because they’re not parties to the contract.

So to put it in terms of your cake analogy, if you have proprietary eggs and free flour, you can ship them together in a box with instructions on how to mix them to make a cake. Or if you own the rights to the eggs, you can combine them and release them as a free cake. Conversely, you can go to the guy (or guys) who wrote the free flour and ask them to license their code to you under a proprietary license so you can make a proprietary cake. The only thing you can’t do is put the free flour into a proprietary cake without the permission of the guy who made the flour. Or, more specifically, you couldn’t make a cake out of the flour and then get a patent on the cake to prevent anyone but you from eating it—because that effectively expropriates the “intellectual property” of the guy who made the flour.

The new version of the GPL is about closing loopholes that might allow someone to effectively put free eggs into proprietary cakes. TiVo found such a loophole when they sold devices that had embedded Linux without making the source code available. Many people believe the Microsoft-Novell agreement will wind up being another such loophole, because it could be a prelude to Microsoft suing other companies that haven’t paid Microsoft for the privilege of using Linux. These really are loopholes, because they lead to free software being transformed into unfree software: in the former case because you can’t get the source code, in the latter case because you can’t use it without paying Microsoft for the privilege. It has nothing to do with stopping collaboration, and everything to do with protecting the rights of free software developer to prevent others from expropriating their intellectual creations.

“So we seem to be left with a semantic dispute about the meaning of the term “collaboration.” The GPL restricts those forms of collaboration that involve incorporating free software into a proprietary product.”

Another, very simple way to say this is that the GPl does not permit someone to steal code from the GPL commons..

“So we seem to be left with a semantic dispute about the meaning of the term “collaboration.” The GPL restricts those forms of collaboration that involve incorporating free software into a proprietary product.”

Another, very simple way to say this is that the GPl does not permit someone to steal code from the GPL commons..

Not to put words into Mark’s mouth, but I think that Tim’s last comment (April 14 3:47 PM) does a good job of characterizing some of the “loopholes” that the GPL 3 is attempting to plug. And as I’ve argued on the TLF in the recent past, I think that the FSF risks alienating itself in the marketplace by doing this. It’s a debatable point, obviously, because I’m making a prediction. But I think it is a fair argument to at least assert.

And not to put words into the rest of my fellow TLF bloggers, but what is neither good nor fair is enigma_foundry’s credibility attack. Obviously one can analyze the motives behind arguments, but here at the TLF we’ve done a pretty good job of not targeting individual bloggers or commenters, just their message.

Instead, it is my observation that TLF bloggers and commenters are held strictly liable for their argument’s facts and assumptions. It’s a great intellectual debate. And it just has to be this way for administrability purposes, or else the dialog disintegrates into blather like it does on slashdot sometimes. Enigma_foundry’s comment, which cast aspersions over a person’s place of employ, is better suited toward slashdot, where it is common to hide behind a pseudonym.

Impugning credibility is a rhetorical technique that has an appropriate time and place. But when it comes to the TLF’s own bloggers and commenters, let’s rise above the fray.

Let me second Braden’s comment: I see no reason to believe Mark isn’t taking the positions he takes in good faith, and I don’t think anything is served by impugning Mark’s motives. If Mark’s arguments are wrong (and I think some of them are) it’s much more productive to explain what’s wrong with his arguments than to speculate on who signs his paychecks.

Not to put words into Mark’s mouth, but I think that Tim’s last comment (April 14 3:47 PM) does a good job of characterizing some of the “loopholes” that the GPL 3 is attempting to plug. And as I’ve argued on the TLF in the recent past, I think that the FSF risks alienating itself in the marketplace by doing this. It’s a debatable point, obviously, because I’m making a prediction. But I think it is a fair argument to at least assert.

And not to put words into the rest of my fellow TLF bloggers, but what is neither good nor fair is enigma_foundry’s credibility attack. Obviously one can analyze the motives behind arguments, but here at the TLF we’ve done a pretty good job of not targeting individual bloggers or commenters, just their message.

Instead, it is my observation that TLF bloggers and commenters are held strictly liable for their argument’s facts and assumptions. It’s a great intellectual debate. And it just has to be this way for administrability purposes, or else the dialog disintegrates into blather like it does on slashdot sometimes. Enigma_foundry’s comment, which cast aspersions over a person’s place of employ, is better suited toward slashdot, where it is common to hide behind a pseudonym.

Impugning credibility is a rhetorical technique that has an appropriate time and place. But when it comes to the TLF’s own bloggers and commenters, let’s rise above the fray.

Let me second Braden’s comment: I see no reason to believe Mark isn’t taking the positions he takes in good faith, and I don’t think anything is served by impugning Mark’s motives. If Mark’s arguments are wrong (and I think some of them are) it’s much more productive to explain what’s wrong with his arguments than to speculate on who signs his paychecks.

Well, as tim had noted, the use of the term loophole was misleading, and I found a pattern of Mark’s not understanding his opponents points, for example in his qutations from which he imputes that he, a paid by Microsoft shill has a better understanding of the four freedoms of the GPL than the clearly stated interpretations of the GPL found on the FSF website.

In particular, Mark seems intent on painting the FSF as a anti-freedom organization, which is really just a smear. Mark should apologize to the FSF.

I suppose, of course, Mark really raises the level of the debate when he says in his highly intellectual style:

Wow, you remind me of my ex-girlfriend. No matter what I say, you only hear what you want to hear, and refuse to ever admit when you’re wrong.

I would simply remind everyone, including Braden Cox, that when Richard Bennett said the following, not one of the TLF asked him to make an apology:

enigma_foundry, you’re a scurrilous rat-bastard. You sit there hiding behind your pseudonym taking shots at real people who use real names and have real biographies. Who do you work for and where does their money come from, cowardly pseudonymous wuss?

Braden Cox’ arguments either make sense or they don’t, and his employer’s sources of funding have shit to do with it.

Posted by: Richard Bennett on April 4, 2007 4:56 AM

So, is it worse to use expletives without argument, or to seek to understand the linkages, some of them economic, which seem to influence certain people’s views of the facts?

Taking note of funding, especially when people make a claim to be “analytically removed” from the debate is fair.

I will leave with Doug Lay’s excellent comment:

Engima did not say that Braden’s arguments have no value. He questioned Braden’s claim to be “analytically removed” from the debate. I think that’s completely fair. Braden may not have an emotional stake in the debate, but he sure has a professional stake. Claiming some sort of 3rd-party objectivity just doesn’t wash. My recommendation for ACT folks, which you can take or leave, is to be up-front about representing the interests of commercial software companies, and to explain why these companies and their views are important and have value.

Posted by: Doug Lay on April 4, 2007 11:14 AM

Well, as tim had noted, the use of the term loophole was misleading, and I found a pattern of Mark’s not understanding his opponents points, for example in his qutations from which he imputes that he, a paid by Microsoft shill has a better understanding of the four freedoms of the GPL than the clearly stated interpretations of the GPL found on the FSF website.

In particular, Mark seems intent on painting the FSF as a anti-freedom organization, which is really just a smear. Mark should apologize to the FSF.

I suppose, of course, Mark really raises the level of the debate when he says in his highly intellectual style:

Wow, you remind me of my ex-girlfriend. No matter what I say, you only hear what you want to hear, and refuse to ever admit when you’re wrong.

I would simply remind everyone, including Braden Cox, that when Richard Bennett said the following, not one of the TLF asked him to make an apology:

enigma_foundry, you’re a scurrilous rat-bastard. You sit there hiding behind your pseudonym taking shots at real people who use real names and have real biographies. Who do you work for and where does their money come from, cowardly pseudonymous wuss?

Braden Cox’ arguments either make sense or they don’t, and his employer’s sources of funding have shit to do with it.

Posted by: Richard Bennett on April 4, 2007 4:56 AM

So, is it worse to use expletives without argument, or to seek to understand the linkages, some of them economic, which seem to influence certain people’s views of the facts?

Taking note of funding, especially when people make a claim to be “analytically removed” from the debate is fair.

I will leave with Doug Lay’s excellent comment:

Engima did not say that Braden’s arguments have no value. He questioned Braden’s claim to be “analytically removed” from the debate. I think that’s completely fair. Braden may not have an emotional stake in the debate, but he sure has a professional stake. Claiming some sort of 3rd-party objectivity just doesn’t wash. My recommendation for ACT folks, which you can take or leave, is to be up-front about representing the interests of commercial software companies, and to explain why these companies and their views are important and have value.

Posted by: Doug Lay on April 4, 2007 11:14 AM

Braden, thanks for toning down the debate and getting us back on track.

Tim, I think you’re absolutely right that we’re now just debating the meaning of "collaboration."  Which is probably the reason we’ve been talking past each other for the past couple weeks.

I think your description of the types of collaboration allowed/not allowed are generally accurate, but I would offer a few addendums, etc, that I believe are important. You also got me thinking about some things I missed in our earlier discussion.

In other words, the only types of "collaboration" it restricts are the ones that involve companies putting other peoples’ code into their products without permission.

I think that is generally right, but with a few important exceptions:

1. Free Software developers cannot integrate proprietary code (even with consent) into their software.  This level of integration isn’t always necessary, as we’ve discussed various workarounds, but there are many scenarios where this would be really helpful. One of the key areas of criticism for Linux and other Free Software programs is usability.  You can’t have Apple-type hit-the-power-button-and-create-a-video simplicity with a mixed source product (including GPL).  Your point that "It does not restrict forms of collaboration that involve incorporating proprietary software into a free software product and releasing the whole thing under the GPL" is accurate, but the important part of the sentence is "releasing the whole thing under the GPL" which means that the proprietary code isn’t proprietary anymore.  So the consent of the proprietary developer must go beyond licensing the code to the Free Software developer and into giving up his rights to keep that code proprietary.
2. The restrictions of the GPL DEFINITELY don’t prevent interoperability ("work better together") between FS and Proprietary, but they do limit some avenues.  For proprietary formats, protocols, and technologies, GPL developers are often forced to reverse engineer rather than simply licensing reusable code.   Since some of these things can also be covered by software patents (whether you believe in them or not, they still exist), the new patent language could limit some efforts at interop with those technologies.
3. The ingredients+directions in a box strategy works both ways. Microsoft could ship GPL binaries in a separate directory from Windows (and for all I know they already do) and then give user the directions for integrating them on their home/business computers. I doubt many GPL developers really want Microsoft using their code in anyway, but this is allowed under the GPL.

Actually, your point about dual-licensing schemes (ie MySQL, etc) is a really important  one that I hadn’t thought about in relationship to our discussion.  Proprietary developers can go to a developer of GPL-based code and license their technology under an alternative proprietary license (if the GPL developer is interested in providing one).  This would allow the proprietary developer to integrate that (formerly GPL) code seamlessly into their proprietary software package.  Unfortuantely, I’m not sure this is an avenue that would be open to those creating GPL-based products. 

There are a couple points, however, where you’re a bit further off the mark.

The GPL gives Linus Torvalds and his fellow kernel developers the option of restricting the use of proprietary drivers in the Linux kernel, but if they want to permit proprietary drivers, there’s nothing that Eben Moglen and Richard Stallman can do about it, because they’re not parties to the contract.

The GPL technically is a copyright license, not a contract (although there are questions, especially in v3 if it would hold up as a license).  Therefore you’re interpretation is wrong; Stallman and Moglen don’t need to be either the licensee or the licensor.  It is Stallman and Moglen who have all the power in this equation.  The FSF can (and has) file lawsuits against firms that it believes are misusing GPL code, even if they are not the copyright holder.  In the end, the courts will be the final arbiter of who is right.

Second, I think you misunderstand why the FSF is pissed at Tivo.  Tivo does provide all the changes to Linux it makes for use on the Tivo machine in accordance with the GPL. The Tivoization problem is a different thing entirely.  As explained on Wikipedia:

So while TiVo has complied with the GPL v.2 requirement to release the source code for others to modify, any modified software will not run on TiVo’s hardware.

So, the problem for Stallman and the FSF with Tivo, is that users can’t hack the source code and run it on the Tivo hardware.  For me, this seems more of an ideological point than a practical one of protecting the commons.  Tivo does give back to the commons in accordance with the GPLv2.  But, their business model is such that they are selling a service to you and subsidizing the hardware.  It is similar to the game console markets (both Playstation and XBox are sold at big losses to Sony and MS). The anti-Tivo provision, therefore, would preclude this kind of business model for GPLv3 software. It may not be a terrible thing, but the embedded devices market was a very promising area for open source software.

Look forward to your thoughts on these issues, Tim. Glad we figured out we’re a lot closer to agreement than the rhetoric would suggest too.

Braden, thanks for toning down the debate and getting us back on track.

Tim, I think you’re absolutely right that we’re now just debating the meaning of "collaboration."  Which is probably the reason we’ve been talking past each other for the past couple weeks.

I think your description of the types of collaboration allowed/not allowed are generally accurate, but I would offer a few addendums, etc, that I believe are important. You also got me thinking about some things I missed in our earlier discussion.

In other words, the only types of "collaboration" it restricts are the ones that involve companies putting other peoples’ code into their products without permission.

I think that is generally right, but with a few important exceptions:

<ol><li>Free Software developers cannot integrate proprietary code (even with consent) into their software.  This level of integration isn’t always necessary, as we’ve discussed various workarounds, but there are many scenarios where this would be really helpful. One of the key areas of criticism for Linux and other Free Software programs is usability.  You can’t have Apple-type hit-the-power-button-and-create-a-video simplicity with a mixed source product (including GPL).  Your point that "It does not restrict forms of collaboration that involve incorporating proprietary software into a free software product and releasing the whole thing under the GPL" is accurate, but the important part of the sentence is "releasing the whole thing under the GPL" which means that the proprietary code isn’t proprietary anymore.  So the consent of the proprietary developer must go beyond licensing the code to the Free Software developer and into giving up his rights to keep that code proprietary.</li>
<li>The restrictions of the GPL DEFINITELY don’t prevent interoperability ("work better together") between FS and Proprietary, but they do limit some avenues.  For proprietary formats, protocols, and technologies, GPL developers are often forced to reverse engineer rather than simply licensing reusable code.   Since some of these things can also be covered by software patents (whether you believe in them or not, they still exist), the new patent language could limit some efforts at interop with those technologies. </li>
<li>The ingredients+directions in a box strategy works both ways. Microsoft could ship GPL binaries in a separate directory from Windows (and for all I know they already do) and then give user the directions for integrating them on their home/business computers. I doubt many GPL developers really want Microsoft using their code in anyway, but this is allowed under the GPL. </li></ol>

Actually, your point about dual-licensing schemes (ie MySQL, etc) is a really important  one that I hadn’t thought about in relationship to our discussion.  Proprietary developers can go to a developer of GPL-based code and license their technology under an alternative proprietary license (if the GPL developer is interested in providing one).  This would allow the proprietary developer to integrate that (formerly GPL) code seamlessly into their proprietary software package.  Unfortuantely, I’m not sure this is an avenue that would be open to those creating GPL-based products. 

There are a couple points, however, where you’re a bit further off the mark.

The GPL gives Linus Torvalds and his fellow kernel developers the option of restricting the use of proprietary drivers in the Linux kernel, but if they want to permit proprietary drivers, there’s nothing that Eben Moglen and Richard Stallman can do about it, because they’re not parties to the contract.

The GPL technically is a copyright license, not a contract (although there are questions, especially in v3 if it would hold up as a license).  Therefore you’re interpretation is wrong; Stallman and Moglen don’t need to be either the licensee or the licensor.  It is Stallman and Moglen who have all the power in this equation.  The FSF can (and has) file lawsuits against firms that it believes are misusing GPL code, even if they are not the copyright holder.  In the end, the courts will be the final arbiter of who is right.

Second, I think you misunderstand why the FSF is pissed at Tivo.  Tivo does provide all the changes to Linux it makes for use on the Tivo machine in accordance with the GPL. The Tivoization problem is a different thing entirely.  As explained on Wikipedia:

So while TiVo has complied with the GPL v.2 requirement to release the source code for others to modify, any modified software will not run on TiVo’s hardware.

So, the problem for Stallman and the FSF with Tivo, is that users can’t hack the source code and run it on the Tivo hardware.  For me, this seems more of an ideological point than a practical one of protecting the commons.  Tivo does give back to the commons in accordance with the GPLv2.  But, their business model is such that they are selling a service to you and subsidizing the hardware.  It is similar to the game console markets (both Playstation and XBox are sold at big losses to Sony and MS). The anti-Tivo provision, therefore, would preclude this kind of business model for GPLv3 software. It may not be a terrible thing, but the embedded devices market was a very promising area for open source software.

Look forward to your thoughts on these issues, Tim. Glad we figured out we’re a lot closer to agreement than the rhetoric would suggest too.

Enigma,

I have never said that the FSF is anti-Freedom. I have pointed out that that the FSF has a very specific definition of “freedom” (the Four Freedoms, 0-3), which not everyone would agree with. I’ve never tried to interpret the Four Freedoms in anyway other than in which they were written.

To be perfectly honest, I can’t think of a single thing I’ve written, which the FSF would actually disagree with. Perhaps there are a couple words issues, but my facts are all well researched and backed up with the links.

I know it’s odd for a blogger-type, but I actually spend a lot of time researching before I write. I want to make sure that I have my facts right and backed up with actual references before heading into the lions den. I know that people like you will always look skeptically at what I write, because of where I work. That is your right, and it wise to understand the point of view of author. But it isn’t wise or productive to argue that “Mark says the sky is blue, therefore it must be purple” simply because of where I work.

If you want to refute my facts or arguments, do so with your own. I take the time to understand the subject before I write and back up my arguments. I would really appreciate it if you would do the same. Don’t just fall back to calling me a “Microsoft Shill.”

Enigma,

I have never said that the FSF is anti-Freedom. I have pointed out that that the FSF has a very specific definition of “freedom” (the Four Freedoms, 0-3), which not everyone would agree with. I’ve never tried to interpret the Four Freedoms in anyway other than in which they were written.

To be perfectly honest, I can’t think of a single thing I’ve written, which the FSF would actually disagree with. Perhaps there are a couple words issues, but my facts are all well researched and backed up with the links.

I know it’s odd for a blogger-type, but I actually spend a lot of time researching before I write. I want to make sure that I have my facts right and backed up with actual references before heading into the lions den. I know that people like you will always look skeptically at what I write, because of where I work. That is your right, and it wise to understand the point of view of author. But it isn’t wise or productive to argue that “Mark says the sky is blue, therefore it must be purple” simply because of where I work.

If you want to refute my facts or arguments, do so with your own. I take the time to understand the subject before I write and back up my arguments. I would really appreciate it if you would do the same. Don’t just fall back to calling me a “Microsoft Shill.”

Your point that “It does not restrict forms of collaboration that involve incorporating proprietary software into a free software product and releasing the whole thing under the GPL” is accurate, but the important part of the sentence is “releasing the whole thing under the GPL” which means that the proprietary code isn’t proprietary anymore. So the consent of the proprietary developer must go beyond licensing the code to the Free Software developer and into giving up his rights to keep that code proprietary.

This is a completely symmetrical situation. You’re right: when you add proprietary code to a GPLed program, the proprietary parts cease to be proprietary. By the same token, if you add free software to a proprietary product, the free parts cease to be free. So why do you object to the GPL, which requires that free software remain free, but not to (say) Microsoft’s EULA, which insists that proprietary software remain proprietary?

For proprietary formats, protocols, and technologies, GPL developers are often forced to reverse engineer rather than simply licensing reusable code.

Free software developers can’t re-use proprietary code unless the proprietary vendor is willing to license it under a free license. Conversely, proprietary software developers can’t re-use free software unless the author is willing to license it under a proprietary license. I don’t see how this is a flaw of the GPL rather than proprietary software licenses.

The FSF can (and has) file lawsuits against firms that it believes are misusing GPL code, even if they are not the copyright holder.

Really? Can you give me an example?

Your point that “It does not restrict forms of collaboration that involve incorporating proprietary software into a free software product and releasing the whole thing under the GPL” is accurate, but the important part of the sentence is “releasing the whole thing under the GPL” which means that the proprietary code isn’t proprietary anymore. So the consent of the proprietary developer must go beyond licensing the code to the Free Software developer and into giving up his rights to keep that code proprietary.

This is a completely symmetrical situation. You’re right: when you add proprietary code to a GPLed program, the proprietary parts cease to be proprietary. By the same token, if you add free software to a proprietary product, the free parts cease to be free. So why do you object to the GPL, which requires that free software remain free, but not to (say) Microsoft’s EULA, which insists that proprietary software remain proprietary?

For proprietary formats, protocols, and technologies, GPL developers are often forced to reverse engineer rather than simply licensing reusable code.

Free software developers can’t re-use proprietary code unless the proprietary vendor is willing to license it under a free license. Conversely, proprietary software developers can’t re-use free software unless the author is willing to license it under a proprietary license. I don’t see how this is a flaw of the GPL rather than proprietary software licenses.

The FSF can (and has) file lawsuits against firms that it believes are misusing GPL code, even if they are not the copyright holder.

Really? Can you give me an example?

Free Software developers cannot integrate proprietary code (even with consent) into their software. This level of integration isn’t always necessary, as we’ve discussed various workarounds, but there are many scenarios where this would be really helpful. One of the key areas of criticism for Linux and other Free Software programs is usability. You can’t have Apple-type hit-the-power-button-and-create-a-video simplicity with a mixed source product (including GPL).

Well, there is no reason at all that you can’t have that ‘Apple’ experience, with a modern distro. In fact I had set up my own computer using SuSE Linux (TM) 10.0, and I would say that once the OS is installed, it is a very Apple-like experience, even down to the liquid-like sliders and progress bars, if you want that. Digital camera’s when plugged in just get recognized, automagically. My HP printer scanner got recognized and configured correctly the first time I had plugged it in. My old serial port Wacom Art Pad worked with SuSE, and it is not even supported by Windows XP 64 bit edition. I have to download a special sound driver with Windows, as I got no sound at all from the drivers that were on the Windows CD.

So with Linux (TM), all of my hardware is supported, with only one download (For 3D accelerated driver) for my NVIDIA card, but if I didn’t want that, I could have a totally working system with only what SuSe shipped on their DVD.

Whereas under Windows XP, a supposedly ‘integrated’ system, I had to download special software, or use a vendor supplied cdrom for each of the following:

(1) HP PSC 1210 printer/scanner/copier (2) Samsung ML-2010 Printer (3) Sound Card (CMI 9739) (4) NVidia graphics Card (5) Digital Camara

and then my Wacom ArtPad still will not work!

Anyone who installs SuSE will, I expect have a similar experience, as almost all the drivers ship on the SuSE disk.

Recall that there are very few pre-insalled computers sold with Linux installed, so users still have to install Linux. One of the chief reason this is the case is Microsoft’s anti-competitive actions requiring PC makers to pay a Windows tax for every PC made, whether or not it shipped with Windows. But, outside of Microsoft’s attempts to interfere with the growth of Linux (TM) there is no reason at all that open source or GPL software is incompatible with ease of use.

As Dell and other large PC manufacturers begin to ship with Linux (TM) we should see the ease of use of Linux continue to improve, of course.

I know it’s odd for a blogger-type, but I actually spend a lot of time researching before I write. I want to make sure that I have my facts right and backed up with actual references before heading into the lions den. I know that people like you will always look skeptically at what I write, because of where I work. That is your right, and it wise to understand the point of view of author.

I have never said they weren’t well researched, I have concerns with the distortions. In particular I agree with Tim’s observation that the situation with the GPL and the MS EULA is symetrical. I would like to here you respond to that.

I would be very interested to hear your remarks on my following observation, that all that the FSF is trying to do with the GPL 3.0 is prevent software that is part of the GPL commons from being stolen. Theft is a serious crime, and that is basically what TIVO has done, steal from the GPL commons without giving back. With that being said, I don’t see why anyone would expect the FSF to do anything different than to try to plug the REAL loopholes in the GPL 2, to prevent the theft of code from the GPL commons.

I think the reason that Microsoft is so upset is that the GPL is a lever which is in the process of radically demonetizing certain portions of the software industry. It is a classical example of Schumpeterian creative destruction. In such a process, almost everyone benefits, from the lower cost of the product. There will be one loser though, the particular businesses that are the target of the crative destruction. It just so happens that the shrink-wrapped software manufacturering is being destroyed in this process. Of course they will complain and ask for all sorts of special favors, as they go down. But go down they will.

Mark says:

“Enigma,

I have never said that the FSF is anti-Freedom.”

But then, on his blog, I find that Mark has made the following statment:

“The GPL is designed to force anyone who uses that software to accept the ideology of the FSF either for moral or pragmatic reasons.”

Hmmm…The lower quote sure sounds like you are trying to paint the FSF as anti-freedom. Please do explain, Mark.

First, let me say that I don’t object to the GPL. I don’t subscribe the FSF’s ideology, but I certainly understand the purpose of the GPL and its merits. At the same time, I think it comes with significant limtations. It’s great for some projects, but not all.

I don’t think you’re assertion that “if you add free software to a proprietary product, the free parts cease to be free” is really accurate. Unless you’re making the ideological point that any concessiont to proprietary software is to relinquish your freedom, that is. Most proprietary licenses do not have the reciporical nature of the GPL that could/would force a change in the license of software that shipped with it. In fact, Apple does integrate BSD code as the basis for its new operating system and this hasn’t changed that BSD code into proprietary code.

On the issue of code reuse, I think that is a fair point (although both could integrate with some open source licenses).

Just looked up the DLink example I was thinking about, and it was actually the gpl-violations.org project (a FSF-related org) that sued DLink over violations of the GPL in a German court. http://www.linux.com/article.pl?sid=06/09/24/1252212

I’m pretty sure that that the FSF itself has also threatened or launched lawsuits against violators in the past too. Will do some more looking around in the morning, but if anyone else has examples feel free to toss the out.

Enigma, thanks for your very substantive comment. I’m on the verge of passing out tonight and want to give it the response it deserves, so I’ll get back to you on all the points tomorrow.

I definitely agree with you that SuSe is getting really good. Have you used Ubuntu? What’s your experience there? Personally, I’m excited about experimenting with their new distros once I get Morgan to clean off one of our old PCs.

I do want to make sure you know that I make a distinction between Apple usability and Windows usability, especially when it comes to first use. The Apple experience is nearly seamless, like plugging in a telephone, it just works. Rare that you ever have to download drivers (because they control all the hardware). That is something that you’ll always have difficulty reaching with mixed source solutions because you need to do at least some setup at the start (Microsoft, well, they have other issues in this area). Not a problem for you and me, but grandma… well, she still wants it to work like the yellow rotary dial she had in the 60s.

Free Software developers cannot integrate proprietary code (even with consent) into their software. This level of integration isn’t always necessary, as we’ve discussed various workarounds, but there are many scenarios where this would be really helpful. One of the key areas of criticism for Linux and other Free Software programs is usability. You can’t have Apple-type hit-the-power-button-and-create-a-video simplicity with a mixed source product (including GPL).

Well, there is no reason at all that you can’t have that ‘Apple’ experience, with a modern distro. In fact I had set up my own computer using SuSE Linux (TM) 10.0, and I would say that once the OS is installed, it is a very Apple-like experience, even down to the liquid-like sliders and progress bars, if you want that. Digital camera’s when plugged in just get recognized, automagically. My HP printer scanner got recognized and configured correctly the first time I had plugged it in. My old serial port Wacom Art Pad worked with SuSE, and it is not even supported by Windows XP 64 bit edition. I have to download a special sound driver with Windows, as I got no sound at all from the drivers that were on the Windows CD.

So with Linux (TM), all of my hardware is supported, with only one download (For 3D accelerated driver) for my NVIDIA card, but if I didn’t want that, I could have a totally working system with only what SuSe shipped on their DVD.

Whereas under Windows XP, a supposedly ‘integrated’ system, I had to download special software, or use a vendor supplied cdrom for each of the following:

(1) HP PSC 1210 printer/scanner/copier
(2) Samsung ML-2010 Printer
(3) Sound Card (CMI 9739)
(4) NVidia graphics Card
(5) Digital Camara

and then my Wacom ArtPad still will not work!

Anyone who installs SuSE will, I expect have a similar experience, as almost all the drivers ship on the SuSE disk.

Recall that there are very few pre-insalled computers sold with Linux installed, so users still have to install Linux. One of the chief reason this is the case is Microsoft’s anti-competitive actions requiring PC makers to pay a Windows tax for every PC made, whether or not it shipped with Windows. But, outside of Microsoft’s attempts to interfere with the growth of Linux (TM) there is no reason at all that open source or GPL software is incompatible with ease of use.

As Dell and other large PC manufacturers begin to ship with Linux (TM) we should see the ease of use of Linux continue to improve, of course.

I know it’s odd for a blogger-type, but I actually spend a lot of time researching before I write. I want to make sure that I have my facts right and backed up with actual references before heading into the lions den. I know that people like you will always look skeptically at what I write, because of where I work. That is your right, and it wise to understand the point of view of author.

I have never said they weren’t well researched, I have concerns with the distortions. In particular I agree with Tim’s observation that the situation with the GPL and the MS EULA is symetrical. I would like to here you respond to that.

I would be very interested to hear your remarks on my following observation, that all that the FSF is trying to do with the GPL 3.0 is prevent software that is part of the GPL commons from being stolen. Theft is a serious crime, and that is basically what TIVO has done, steal from the GPL commons without giving back. With that being said, I don’t see why anyone would expect the FSF to do anything different than to try to plug the REAL loopholes in the GPL 2, to prevent the theft of code from the GPL commons.

I think the reason that Microsoft is so upset is that the GPL is a lever which is in the process of radically demonetizing certain portions of the software industry. It is a classical example of Schumpeterian creative destruction. In such a process, almost everyone benefits, from the lower cost of the product. There will be one loser though, the particular businesses that are the target of the crative destruction. It just so happens that the shrink-wrapped software manufacturering is being destroyed in this process. Of course they will complain and ask for all sorts of special favors, as they go down. But go down they will.

Mark says:

“Enigma,

I have never said that the FSF is anti-Freedom.”

But then, on his blog, I find that Mark has made the following statment:

“The GPL is designed to force anyone who uses that software to accept the ideology of the FSF either for moral or pragmatic reasons.”

Hmmm…The lower quote sure sounds like you are trying to paint the FSF as anti-freedom. Please do explain, Mark.

I don’t think you’re assertion that “if you add free software to a proprietary product, the free parts cease to be free” is really accurate. Unless you’re making the ideological point that any concession to proprietary software is to relinquish your freedom, that is. Most proprietary licenses do not have the reciporical nature of the GPL that could/would force a change in the license of software that shipped with it. In fact, Apple does integrate BSD code as the basis for its new operating system and this hasn’t changed that BSD code into proprietary code.

In this case the BSD code is transformed into non-free software in the process, is it not? Users do not have the right to examine, modify, and re-distribute the BSD-derived parts of Apple’s proprietary software. There’s nothing wrong with that—the BSD license explicitly allows it—but it self-evidently does cause “a change in the license of software that shipped with it.”

I’m not sure what’s ideological about the observation that incorporating free software into a proprietary product leads to a reduction of users’ freedom with respect to the new software. Maybe you don’t value the freedom to examine, modify, and redistribute the software you use, which is perfectly fine. However, a lot of software developers do care about those freedoms, and they often use the GPL as a tool to preserve them for their users and the users of derivative software.

I look forward to examples of FSF lawsuits, as I’m still waiting for an example of how the FSF “actively attempts to limit collaboration between proprietary and free software communities.”

First, let me say that I don’t object to the GPL. I don’t subscribe the FSF’s ideology, but I certainly understand the purpose of the GPL and its merits. At the same time, I think it comes with significant limtations. It’s great for some projects, but not all.

I don’t think you’re assertion that “if you add free software to a proprietary product, the free parts cease to be free” is really accurate. Unless you’re making the ideological point that any concessiont to proprietary software is to relinquish your freedom, that is. Most proprietary licenses do not have the reciporical nature of the GPL that could/would force a change in the license of software that shipped with it. In fact, Apple does integrate BSD code as the basis for its new operating system and this hasn’t changed that BSD code into proprietary code.

On the issue of code reuse, I think that is a fair point (although both could integrate with some open source licenses).

Just looked up the DLink example I was thinking about, and it was actually the gpl-violations.org project (a FSF-related org) that sued DLink over violations of the GPL in a German court.
http://www.linux.com/article.pl?sid=06/09/24/12…

I’m pretty sure that that the FSF itself has also threatened or launched lawsuits against violators in the past too. Will do some more looking around in the morning, but if anyone else has examples feel free to toss the out.

Also, gpl-violations.org confirms my point that only the copyright holder (or those to whom the copyright holder has conferred authority) can go to court to enforce the GPL:

The founder, Harald Welte, started to get active with GPL enforcement in late 2003, where he discovered the first bunch of companies violating the GPL in software he wrote for the netfilter/iptables project… Over time, some other Linux kernel developers have transferred their rights in a fiduciary license agreement to enable the successful gpl-violations.org project to enforce the GPL in cases where no code originally written by Harald Welte was used / infringed upon.

As far as I can see, the FSF had no direct role whatsoever.

Enigma, thanks for your very substantive comment. I’m on the verge of passing out tonight and want to give it the response it deserves, so I’ll get back to you on all the points tomorrow.

I definitely agree with you that SuSe is getting really good. Have you used Ubuntu? What’s your experience there? Personally, I’m excited about experimenting with their new distros once I get Morgan to clean off one of our old PCs.

I do want to make sure you know that I make a distinction between Apple usability and Windows usability, especially when it comes to first use. The Apple experience is nearly seamless, like plugging in a telephone, it just works. Rare that you ever have to download drivers (because they control all the hardware). That is something that you’ll always have difficulty reaching with mixed source solutions because you need to do at least some setup at the start (Microsoft, well, they have other issues in this area). Not a problem for you and me, but grandma… well, she still wants it to work like the yellow rotary dial she had in the 60s.

I don’t think you’re assertion that “if you add free software to a proprietary product, the free parts cease to be free” is really accurate. Unless you’re making the ideological point that any concession to proprietary software is to relinquish your freedom, that is. Most proprietary licenses do not have the reciporical nature of the GPL that could/would force a change in the license of software that shipped with it. In fact, Apple does integrate BSD code as the basis for its new operating system and this hasn’t changed that BSD code into proprietary code.

In this case the BSD code is transformed into non-free software in the process, is it not? Users do not have the right to examine, modify, and re-distribute the BSD-derived parts of Apple’s proprietary software. There’s nothing wrong with that—the BSD license explicitly allows it—but it self-evidently does cause “a change in the license of software that shipped with it.”

I’m not sure what’s ideological about the observation that incorporating free software into a proprietary product leads to a reduction of users’ freedom with respect to the new software. Maybe you don’t value the freedom to examine, modify, and redistribute the software you use, which is perfectly fine. However, a lot of software developers do care about those freedoms, and they often use the GPL as a tool to preserve them for their users and the users of derivative software.

I look forward to examples of FSF lawsuits, as I’m still waiting for an example of how the FSF “actively attempts to limit collaboration between proprietary and free software communities.”

Also, gpl-violations.org confirms my point that only the copyright holder (or those to whom the copyright holder has conferred authority) can go to court to enforce the GPL:

The founder, Harald Welte, started to get active with GPL enforcement in late 2003, where he discovered the first bunch of companies violating the GPL in software he wrote for the netfilter/iptables project…

Over time, some other Linux kernel developers have transferred their rights in a fiduciary license agreement to enable the successful gpl-violations.org project to enforce the GPL in cases where no code originally written by Harald Welte was used / infringed upon.

As far as I can see, the FSF had no direct role whatsoever.

Enigma,

Sorry, this day is getting away from me, but wanted to get back to you. Tim, you’re next…

I agree wholeheartedly that the open source operating systems are getting closer and closer to being as user-friendly as Windows or OSX. I know your pain re: XP and device drivers. But, the point I was trying to make yesterday was about out of the box usability (think the Mac vs PC ads). Apple has created a really amazing out-of-the-box experience that everyone wishes they could replicate. They do it through tight control of the system components of course, but the result is pretty close to just "plug it in and use it." When you setup your computer w/ SuSE, you still needed to "set it up" especially if you wanted to use proprietary components. Right?

There are ways around the problems by using the gray areas of the GPL or using the LGPL.  Dell and others shipping Linux-based PCs obviously have spent an awful lot of time figure out the legal issues.  I don’t own one of these PCs, however, so I can’t tell you for sure how close they’ve gotten to an Apple-like experience… but I’m skeptical. As the OLPC situation is proving, even minor segments of proprietary code in the shipping product can be problematic. http://www.thejemreport.com/mambo/content/view/286/ 

You also said that In particular I agree with Tim’s observation that the situation with the GPL and the MS EULA is symetrical. I would like to here you respond to that.

Here is what I wrote back to Tim last night:

I don’t think you’re assertion that "if you add free software to a proprietary product, the free parts cease to be free" is really accurate. Unless you’re making the ideological point that any concession to proprietary software is to relinquish your freedom, that is. Most proprietary licenses do not have the reciprocal nature of the GPL that could/would force a change in the license of software that shipped with it. In fact, Apple does integrate BSD code as the basis for its new operating system and this hasn’t changed that BSD code into proprietary code.

To go further, I can see how on first glance they seem symmetrical. However, there are some significant differences. As Larry Rosen summed up the GPL bargain in his book Open Source Licensing, "You may have this free software on condition hat any derivative works that you create from it and distribute must be licensed under the same license." Proprietary licenses do not typically have this reciprocity clause nor do they guarantee sublicensibility. Proprietary software like Windows and OSX ship under EULA’s that do not override the licenses of the open source software that is included. Apache, BSD, and other open source licenses allow this kind of combination, but protect that code form irrevocably being pulled out of the commons (although the BSD and some others do not protect derivative works).

You next bit was:

I would be very interested to hear your remarks on my following observation, that all that the FSF is trying to do with the GPL 3.0 is prevent software that is part of the GPL commons from being stolen. Theft is a serious crime, and that is basically what TIVO has done, steal from the GPL commons without giving back. With that being said, I don’t see why anyone would expect the FSF to do anything different than to try to plug the REAL loopholes in the GPL 2, to prevent the theft of code from the GPL commons.

It is completely understandable that the FSF would try to use the GPLv3 to prevent theft from the commons, but that is not what the Tivo provision is about. It is often misunderstood, so I understand your confusion on the matter. Here is what I wrote the Tim last night. Please let me know if you still have questions after reading it.

Tivo does provide all the changes to Linux it makes for use on the Tivo machine in accordance with the GPL. The Tivoization problem is a different thing entirely. As explained on Wikipedia:

So while TiVo has complied with the GPL v.2 requirement to release the source code for others to modify, any modified software will not run on TiVo’s hardware.

So, the problem for Stallman and the FSF with Tivo, is that users can’t hack the source code and run it on the Tivo hardware. For me, this seems more of an ideological point than a practical one of protecting the commons. Tivo does give back to the commons in accordance with the GPLv2. But, their business model is such that they are selling a service to you and subsidizing the hardware. It is similar to the game console markets (both Playstation and XBox are sold at big losses to Sony and MS). The anti-Tivo provision, therefore, would preclude this kind of business model for GPLv3 software. It may not be a terrible thing, but the embedded devices market was a very promising area for open source software.

You also wrote that:

I think the reason that Microsoft is so upset is that the GPL is a lever which is in the process of radically demonetizing certain portions of the software industry. It is a classical example of Schumpeterian creative destruction. In such a process, almost everyone benefits, from the lower cost of the product. There will be one loser though, the particular businesses that are the target of the crative destruction. It just so happens that the shrink-wrapped software manufacturering is being destroyed in this process. Of course they will complain and ask for all sorts of special favors, as they go down. But go down they will.

Interesting theory. Personally, I think that the software industry is evolving toward a hybrid-licensing reality where licensing decisions are made based on business case, not by ideology.  The days of all proprietary software are over (nor were they ever really here), but I’m not sold on an all Free Software world either. We are all aware of the potential tragedy of the commons… I see both models surviving for a very long time. But in the near term, open source licensing will only become more important as we head toward some level of equilibrium. But, this sounds like the makings of a much longer post.

In particular I agree with Tim’s observation that the situation with the GPL and the MS EULA is symetrical. I would like to here you respond to that."

Yes, Enigma, I said that or something similar:

“The GPL is designed to force anyone who uses that software to accept the ideology of the FSF either for moral or pragmatic reasons.”

But, that does not say that the FSF is anti-freedom.

You would agree that the GPL is the embodiment of the “Four Freedoms” ideology of Stallman and the Free Software Foundation in a software license, would you not?

Therefore using the GPL requires you to submit to that ideology, either because you believe in it (as a Free Software advocate) or you are using it for pragmatic reasons (the open source community, etc).

The FSF and I don’t agree on what the definition of Freedom is, therefore it is difficult for me to say they are anti-Freedom. That said, I wouldn’t say they are entirely pro-Freedom in my mind either. The GPL places restrictions on what you can do with the software, just like proprietary licenses do. They are different restrictions, but restrictions nonetheless.

Enigma,

Sorry, this day is getting away from me, but wanted to get back to you. Tim, you’re next…

I agree wholeheartedly that the open source operating systems are getting closer and closer to being as user-friendly as Windows or OSX. I know your pain re: XP and device drivers. But, the point I was trying to make yesterday was about out of the box usability (think the Mac vs PC ads). Apple has created a really amazing out-of-the-box experience that everyone wishes they could replicate. They do it through tight control of the system components of course, but the result is pretty close to just "plug it in and use it." When you setup your computer w/ SuSE, you still needed to "set it up" especially if you wanted to use proprietary components. Right?

There are ways around the problems by using the gray areas of the GPL or using the LGPL.  Dell and others shipping Linux-based PCs obviously have spent an awful lot of time figure out the legal issues.  I don’t own one of these PCs, however, so I can’t tell you for sure how close they’ve gotten to an Apple-like experience… but I’m skeptical. As the OLPC situation is proving, even minor segments of proprietary code in the shipping product can be problematic. http://www.thejemreport.com/mambo/content/view/286/ 

You also said that In particular I agree with Tim’s observation that the situation with the GPL and the MS EULA is symetrical. I would like to here you respond to that.

Here is what I wrote back to Tim last night:

I don’t think you’re assertion that "if you add free software to a proprietary product, the free parts cease to be free" is really accurate. Unless you’re making the ideological point that any concession to proprietary software is to relinquish your freedom, that is. Most proprietary licenses do not have the reciprocal nature of the GPL that could/would force a change in the license of software that shipped with it. In fact, Apple does integrate BSD code as the basis for its new operating system and this hasn’t changed that BSD code into proprietary code.

To go further, I can see how on first glance they seem symmetrical. However, there are some significant differences. As Larry Rosen summed up the GPL bargain in his book Open Source Licensing, "You may have this free software on condition hat any derivative works that you create from it and distribute must be licensed under the same license." Proprietary licenses do not typically have this reciprocity clause nor do they guarantee sublicensibility. Proprietary software like Windows and OSX ship under EULA’s that do not override the licenses of the open source software that is included. Apache, BSD, and other open source licenses allow this kind of combination, but protect that code form irrevocably being pulled out of the commons (although the BSD and some others do not protect derivative works).

You next bit was:

I would be very interested to hear your remarks on my following observation, that all that the FSF is trying to do with the GPL 3.0 is prevent software that is part of the GPL commons from being stolen. Theft is a serious crime, and that is basically what TIVO has done, steal from the GPL commons without giving back. With that being said, I don’t see why anyone would expect the FSF to do anything different than to try to plug the REAL loopholes in the GPL 2, to prevent the theft of code from the GPL commons.

It is completely understandable that the FSF would try to use the GPLv3 to prevent theft from the commons, but that is not what the Tivo provision is about. It is often misunderstood, so I understand your confusion on the matter. Here is what I wrote the Tim last night. Please let me know if you still have questions after reading it.

Tivo does provide all the changes to Linux it makes for use on the Tivo machine in accordance with the GPL. The Tivoization problem is a different thing entirely. As explained on Wikipedia:

So while TiVo has complied with the GPL v.2 requirement to release the source code for others to modify, any modified software will not run on TiVo’s hardware.

So, the problem for Stallman and the FSF with Tivo, is that users can’t hack the source code and run it on the Tivo hardware. For me, this seems more of an ideological point than a practical one of protecting the commons. Tivo does give back to the commons in accordance with the GPLv2. But, their business model is such that they are selling a service to you and subsidizing the hardware. It is similar to the game console markets (both Playstation and XBox are sold at big losses to Sony and MS). The anti-Tivo provision, therefore, would preclude this kind of business model for GPLv3 software. It may not be a terrible thing, but the embedded devices market was a very promising area for open source software.

You also wrote that:

I think the reason that Microsoft is so upset is that the GPL is a lever which is in the process of radically demonetizing certain portions of the software industry. It is a classical example of Schumpeterian creative destruction. In such a process, almost everyone benefits, from the lower cost of the product. There will be one loser though, the particular businesses that are the target of the crative destruction. It just so happens that the shrink-wrapped software manufacturering is being destroyed in this process. Of course they will complain and ask for all sorts of special favors, as they go down. But go down they will.

Interesting theory. Personally, I think that the software industry is evolving toward a hybrid-licensing reality where licensing decisions are made based on business case, not by ideology.  The days of all proprietary software are over (nor were they ever really here), but I’m not sold on an all Free Software world either. We are all aware of the potential tragedy of the commons… I see both models surviving for a very long time. But in the near term, open source licensing will only become more important as we head toward some level of equilibrium. But, this sounds like the makings of a much longer post.

In particular I agree with Tim’s observation that the situation with the GPL and the MS EULA is symetrical. I would like to here you respond to that."

Yes, Enigma, I said that or something similar:

“The GPL is designed to force anyone who uses that software to accept the ideology of the FSF either for moral or pragmatic reasons.”

But, that does not say that the FSF is anti-freedom.

You would agree that the GPL is the embodiment of the “Four Freedoms” ideology of Stallman and the Free Software Foundation in a software license, would you not?

Therefore using the GPL requires you to submit to that ideology, either because you believe in it (as a Free Software advocate) or you are using it for pragmatic reasons (the open source community, etc).

The FSF and I don’t agree on what the definition of Freedom is, therefore it is difficult for me to say they are anti-Freedom. That said, I wouldn’t say they are entirely pro-Freedom in my mind either. The GPL places restrictions on what you can do with the software, just like proprietary licenses do. They are different restrictions, but restrictions nonetheless.

Yes, Enigma, I said that or something similar:

“The GPL is designed to force anyone who uses that software to accept the ideology of the FSF either for moral or pragmatic reasons.”

But, that does not say that the FSF is anti-freedom.

It is a fine line to say that the FSF has designed something “to force anyone that uses that software to accept the ideaology” and saying that the FSF is anti-freedom.

Simply put, the complexity of GPL 3.0 mirrors the complexity of the schemes which have been or might be hatched to steal from the GPL commons.

But, back to your first point. It is just plain silly to maintain that in order to abide by the legal requirments of a lisence, your belief system must become aligned with the author of the lisence. Fascists, libertarians, Democrats, Greens, Republicans, anarcho-syndicalists, and Whigs can all abide by the trems of the GPL and maintain any beliefs the desire.

The exciting thing I see in the future is that large for-profit companies will see the benefit of coming together and creating GPL software. But that is really a process that has been occuring in the economy for a long time. It is just the extension of standards making behavior, which has served many industrial sectors well. The GPL is the eans for hardware manufacturers to create certain software, as a standard, and then get on withh the business of selling hardware, that users, using those standards, can use.

If you look at the history of standards, in particular the ANSI, you will see the same types of processes occurred when many of those standards were written.

Yes, Enigma, I said that or something similar:

“The GPL is designed to force anyone who uses that software to accept the ideology of the FSF either for moral or pragmatic reasons.”

But, that does not say that the FSF is anti-freedom.

It is a fine line to say that the FSF has designed something “to force anyone that uses that software to accept the ideaology” and saying that the FSF is anti-freedom.

Simply put, the complexity of GPL 3.0 mirrors the complexity of the schemes which have been or might be hatched to steal from the GPL commons.

But, back to your first point. It is just plain silly to maintain that in order to abide by the legal requirments of a lisence, your belief system must become aligned with the author of the lisence. Fascists, libertarians, Democrats, Greens, Republicans, anarcho-syndicalists, and Whigs can all abide by the trems of the GPL and maintain any beliefs the desire.

The exciting thing I see in the future is that large for-profit companies will see the benefit of coming together and creating GPL software. But that is really a process that has been occuring in the economy for a long time. It is just the extension of standards making behavior, which has served many industrial sectors well. The GPL is the eans for hardware manufacturers to create certain software, as a standard, and then get on withh the business of selling hardware, that users, using those standards, can use.

If you look at the history of standards, in particular the ANSI, you will see the same types of processes occurred when many of those standards were written.

Oh, and a post that is relevant to this discussion, in particular, realizing that selling software as a product is not “Copybot Resistant”

Copybot Resistance

Oh, and a post that is relevant to this discussion, in particular, realizing that selling software as a product is not “Copybot Resistant”

Copybot Resistance

OK, Tim. Here are a few responses to your questions.  You asked:

In this case the BSD code is transformed into non-free software in the process, is it not?

I reponded to this point in my comment to enigma (see above). Proprietary licenses do not inherently change the licenses of the underlying code. The GPL does. Apple does keep the kernel under the BSD license that can be downladed and run separately, it is called the Darwin project. I believe under the BSD terms, Apple could take the code and put release it under a full proprietary license, but they don’t have to and seem to have chosen not to. If the code they were borrowing was under the Apache license, they would be forced to keep it and modfications to it open, but they could still ship it under one EULA. So, no, it does not self-evidently cause a change in the license of the software that is shipped with it.

I think you also misunderstood my "ideological point." I said:

I don’t think you’re assertion that "if you add free software to a proprietary product, the free parts cease to be free" is really accurate. Unless you’re making the ideological point that any concessiont to proprietary software is to relinquish your freedom, that is.

I turns out that your assertion was inaccurate, so you weren’t making the truly "ideological point" that I thought you were making. That point would have been the point that Stallman was making to the Open Office team – that any reliance on proprietary code is to concede your freedom even if all the actual code is GPL.

The FSF lawsuits.

I’m going to do something a little wild for this forum. I’m going to admit that I was wrong. Hopefully, it will set an example for these discussions in the future.

The FSF must be either a licensee or licensor in order to initiate a lawsuit. My sh*thouse law degree is showing its worth…but we both seem to have the problem from time to time.

HOWEVER, while I was technically wrong on that part, my larger point still stands. The GPL claims to be a copyright license, not a contract, and Linus does not have the level of control that you suggest. You wrote that:

But that illustrates an important fact about the GPL: the ultimate power to decide how strictly to interpret its terms is ultimately up to software developers, not the FSF. The GPL gives Linus Torvalds and his fellow kernel developers the option of restricting the use of proprietary drivers in the Linux kernel, but if they want to permit proprietary drivers, there’s nothing that Eben Moglen and Richard Stallman can do about it, because they’re not parties to the contract.

First, the FSF is one of the largest copyright holders in a standard Linux distribution. Not only from things developed by Stallman, but other bits of code that have been signed over to the Free Software Foundation. So, in most cases regarding Linux, the FSF can sue. Mainly, however, they seem content on just threatening to sue as they did against Linksys and Cisco.

http://www.forbes.com/2003/10/14/cz_dl_1014linksys.html

Second, the GPL does not really give developers the "option" of permitting proprietary drivers. The GPL is just not clear in this area (thus all the debate) and the lack of clarity has created some flexibilty for interpretation. This is reinforced by the fact that it has yet to be fully tested legally in the US. Courts could remove that gray area. And the FSF can bring a case in court that will affect ALL GPL code, not just the code they own.

Third, the FSF controls the GPL. If they want to officially clear that up, they can make changes in the next version. Of course Linus has already modified the GPL license for the kernel to remove the automatic upgrade clause, so that won’t automatically create a problem… but it could.

So, the point is: the GPL doesn’t put all the power in the hands of individual software developers as you suggest. The FSF still retains a great deal of power over how the license is interpreted and how GPL software can integrate/interoperate with proprietary software.

OK, Tim. Here are a few responses to your questions.  You asked:

In this case the BSD code is transformed into non-free software in the process, is it not?

I reponded to this point in my comment to enigma (see above). Proprietary licenses do not inherently change the licenses of the underlying code. The GPL does. Apple does keep the kernel under the BSD license that can be downladed and run separately, it is called the Darwin project. I believe under the BSD terms, Apple could take the code and put release it under a full proprietary license, but they don’t have to and seem to have chosen not to. If the code they were borrowing was under the Apache license, they would be forced to keep it and modfications to it open, but they could still ship it under one EULA. So, no, it does not self-evidently cause a change in the license of the software that is shipped with it.

I think you also misunderstood my "ideological point." I said:

I don’t think you’re assertion that "if you add free software to a proprietary product, the free parts cease to be free" is really accurate. Unless you’re making the ideological point that any concessiont to proprietary software is to relinquish your freedom, that is.

I turns out that your assertion was inaccurate, so you weren’t making the truly "ideological point" that I thought you were making. That point would have been the point that Stallman was making to the Open Office team – that any reliance on proprietary code is to concede your freedom even if all the actual code is GPL.

The FSF lawsuits.

I’m going to do something a little wild for this forum. I’m going to admit that I was wrong. Hopefully, it will set an example for these discussions in the future.

The FSF must be either a licensee or licensor in order to initiate a lawsuit. My sh*thouse law degree is showing its worth…but we both seem to have the problem from time to time.

HOWEVER, while I was technically wrong on that part, my larger point still stands. The GPL claims to be a copyright license, not a contract, and Linus does not have the level of control that you suggest. You wrote that:

But that illustrates an important fact about the GPL: the ultimate power to decide how strictly to interpret its terms is ultimately up to software developers, not the FSF. The GPL gives Linus Torvalds and his fellow kernel developers the option of restricting the use of proprietary drivers in the Linux kernel, but if they want to permit proprietary drivers, there’s nothing that Eben Moglen and Richard Stallman can do about it, because they’re not parties to the contract.

First, the FSF is one of the largest copyright holders in a standard Linux distribution. Not only from things developed by Stallman, but other bits of code that have been signed over to the Free Software Foundation. So, in most cases regarding Linux, the FSF can sue. Mainly, however, they seem content on just threatening to sue as they did against Linksys and Cisco.

http://www.forbes.com/2003/10/14/cz_dl_1014linksys.html

Second, the GPL does not really give developers the "option" of permitting proprietary drivers. The GPL is just not clear in this area (thus all the debate) and the lack of clarity has created some flexibilty for interpretation. This is reinforced by the fact that it has yet to be fully tested legally in the US. Courts could remove that gray area. And the FSF can bring a case in court that will affect ALL GPL code, not just the code they own.

Third, the FSF controls the GPL. If they want to officially clear that up, they can make changes in the next version. Of course Linus has already modified the GPL license for the kernel to remove the automatic upgrade clause, so that won’t automatically create a problem… but it could.

So, the point is: the GPL doesn’t put all the power in the hands of individual software developers as you suggest. The FSF still retains a great deal of power over how the license is interpreted and how GPL software can integrate/interoperate with proprietary software.

Enigma,

I didn’t mean that those who sign up for the license must joing the Free Software cause, and I didn’t meant to say that with the word “accept.” It is also why I said for “practical or ideological” reasons.

The exciting thing I see in the future is that large for-profit companies will see the benefit of coming together and creating GPL software. But that is really a process that has been occuring in the economy for a long time. It is just the extension of standards making behavior, which has served many industrial sectors well. The GPL is the eans for hardware manufacturers to create certain software, as a standard, and then get on withh the business of selling hardware, that users, using those standards, can use.

I would argue this is already happening, particularly in regards to Linux. Massive companies like IBM, Sun, and HP have been participating in the effort to improve and evolve the Linux OS. There are definitely parallels to the standards process in this regards.

However, I would suggest that this is not a one-size-fits-all strategy. For-profit companies work together when it makes sense… generally when they need combined resources or when they don’t look at the technology as a competitive differentiator. If the companies are working on really innovative, R&D intensive stuff, it is much harder to make a business case for simply sharing that work with their competitors. Free-rider problems could kill off long-term investment in new technology.

One interesting area that is discussed in the EC FLOSS study is the idea that non IT companies will come together to mutually fund systems development. This is interesting but also limited. A few of my colleagues used to run custom software development shops, with government and large private sector clients (AmEx, etc.). In the private sector, even these companies looked at their software systems as potential competitive differentiators for lowering their costs, improving efficiencies, etc. So, there isn’t always a reason for even non-IT compaines to come together on projects.

Enigma,

I didn’t mean that those who sign up for the license must joing the Free Software cause, and I didn’t meant to say that with the word “accept.” It is also why I said for “practical or ideological” reasons.

The exciting thing I see in the future is that large for-profit companies will see the benefit of coming together and creating GPL software. But that is really a process that has been occuring in the economy for a long time. It is just the extension of standards making behavior, which has served many industrial sectors well. The GPL is the eans for hardware manufacturers to create certain software, as a standard, and then get on withh the business of selling hardware, that users, using those standards, can use.

I would argue this is already happening, particularly in regards to Linux. Massive companies like IBM, Sun, and HP have been participating in the effort to improve and evolve the Linux OS. There are definitely parallels to the standards process in this regards.

However, I would suggest that this is not a one-size-fits-all strategy. For-profit companies work together when it makes sense… generally when they need combined resources or when they don’t look at the technology as a competitive differentiator. If the companies are working on really innovative, R&D; intensive stuff, it is much harder to make a business case for simply sharing that work with their competitors. Free-rider problems could kill off long-term investment in new technology.

One interesting area that is discussed in the EC FLOSS study is the idea that non IT companies will come together to mutually fund systems development. This is interesting but also limited. A few of my colleagues used to run custom software development shops, with government and large private sector clients (AmEx, etc.). In the private sector, even these companies looked at their software systems as potential competitive differentiators for lowering their costs, improving efficiencies, etc. So, there isn’t always a reason for even non-IT compaines to come together on projects.

Peace people

We love you

Peace people

We love you