Re: Rchard Stallman vs Darl McBride
On Sat, Jul 24, 2010 at 1:11 PM, Jon Jensen j...@endpoint.com wrote: Imagine if we had no copyright on software. Sure, people could release only binaries if they wanted. But they'd get copied at will, legally, so there'd be little market for selling them. And the scene would probably largely revert to the way things were back before it was legally decided software (on magnetic media, anyway) could be copyrighted -- and people published source code + binaries fairly freely, either because they were making money off the hardware or because they were part of a community working together to improve software that they all used. I don't think lack of copyright on software would be a good thing, though. I think authors of software should be able, for a limited time, to have exclusive control over how their works are copied. I do think that copyright in general has grown to something beyond what it was intended to be, and its duration should be scaled back to the original 14 years with the option of a 14 year renewal. I think that software copyrights could and ought to be scaled back to an even shorter duration. I also don't think lack of software copyright would have much of an effect on source distribution. In fact, lacking copyright protection, producers of software might be even more reluctant to distribute source code. Lacking copyright, the only way to protect it would be to keep it secret. This is all highly speculative, but there is some historical precedent for it, so we can expect at least in parts of the software world that would happen again, with neither legal nor license requirements compelling it. People using permissive software licenses have been and still are doing things that way. People who didn't want to share would just be more secretive or more creative about trying to hide things. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 7/24/2010 3:13 PM, Levi Pearson wrote: On Sat, Jul 24, 2010 at 1:11 PM, Jon Jensenj...@endpoint.com wrote: SNIP - A bunch of stuff about speculation on copyrights and varying importance of them Food for thought that I found interesting (I mean no more of a statement than that. I'm not endorsing anything here): http://www.ted.com/talks/lang/eng/johanna_blakley_lessons_from_fashion_s_free_culture.html -Tod Hansmann /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 11:36 PM, Aaron Toponce aaron.topo...@gmail.com wrote: I'm surprised this hasn't been brought up, so I'll mention it. Google keeps an internal proprietary derivative of Ubuntu on their employee workstations. They have never released this Goobuntu to the wild, and as such are under no obligation to release their changes. Even if the code is GPL. I'm not sure why you're surprised, as it doesn't really seem relevant. Individual packages within Ubuntu may be GPL-licensed, but the whole distribution doesn't have an overarching license agreement, and I don't think the GPL would work for that either. If they made changes to specific packages that were GPL-licensed and didn't distribute them, that would illustrate a point, but I don't think it's a point that anyone was confused about. At work we have a wireshark plugin for an in-house protocol that we don't distribute. We also developed a couple of decoders for protocols we're helping to develop through IEEE, and we submitted those back upstream. You can find me in the AUTHORS file, though it may not be in a stable release version yet. Sometimes it's useful to have our in-house protocol and the IEEE draft protocols in the same build, but we've got to be careful not to distribute those builds when we're sending out builds with the draft protocol updates to people we're working with. It's a minor annoyance, but annoying nonetheless. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: GPL worldview WAS Re: Rchard Stallman vs Darl McBride
On 07/21/2010 05:18 AM, justin wrote: On Tuesday, July 20, 2010, Michael Torrie torr...@gmail.com wrote: The GPL levels the playing field. IBM's contributions can't be used against them, as they could be with a BSD license. Linux gets better for everyone. I'm not sure I follow. How could IBM's contributions be used against them under a BSD license? Simply put, anything IBM did with a BSD project (and where their code was released under the BSD), could be used in a closed, proprietary way in a product that competes with IBM. Whereas with the GPL, IBM's contributions cannot be closed. If a company chooses to make a superior product based on IBM's contributions under the GPL, IBM is free to incorporate their changes as well (it's only fair after all). Level playing field. IBM works with many open source licenses, but I believe that the GPL is one of the reasons IBM chose (at one time anyway) to invest so heavily in Linux. I do not believe IBM would have invested in the same way in BSD Unix. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: GPL does not tie your hands necessarily - Re: Rchard Stallman vs Darl McBride
On 07/20/2010 10:21 PM, Levi Pearson wrote: If your wrote *all* the code yourself, you absolutely can release only part of it under the GPL. Yes, but that negates the power of using open source building blocks. I assumed some other copyright holders in the mix, either from the beginning or from accepting contributions under the GPL without copyright reassignment. Well incompatible licenses should be a concern no matter what license you are working with. I don't think a GPL library author owes you as a developer anything. So use something that's compatible with your license. If no one can use a library because it's GPL, then it will die and be replaced with something else. I don't see your concern as a genuine problem. If you're working on a proprietary, commercial project, for example, then you should have already taken this into consideration. Also, many GPL'd projects (even ones with multiple copyright holders) are licensed under several different licenese. Yes, but that's GPL-with-exceptions, not vanilla GPL. Structuring things correctly and coming up with the necessary exceptions and all that seem needlessly complicated to me. Well, as I said, writing a special exception to the GPL is not strictly necessary for the copyright holder(s). But a GPL-with-exceptions license can become a vanilla GPL license for the derivative author. Any exception to the GPL need not live past the first generation if you as the developer choose that to be so. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 07/20/2010 10:31 PM, Levi Pearson wrote: Except that 'more liberal' literally means more free. :) But but the tea party says that liberal means socialist fascist extremist baby killer! Doesn't sound more free to me! /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 07/21/2010 12:09 AM, Levi Pearson wrote: I'm not sure why you're surprised, as it doesn't really seem relevant. Individual packages within Ubuntu may be GPL-licensed, but the whole distribution doesn't have an overarching license agreement, and I don't think the GPL would work for that either. If they made changes to specific packages that were GPL-licensed and didn't distribute them, that would illustrate a point, but I don't think it's a point that anyone was confused about. You draw a good point, and while merely an assumption, I think it would be safe to say that Google is making changes to GPL-licensed code, as well as other licensed code. It's heavily speculated that GWS is just a modified version of Apache. The fact that they won't release Goobuntu (or any other custom-built Google appliance), seems to be that they are probably changing GPL, and other licensed cod to fit their own internal needs. Aside from the fact that making these releases, likely doesn't fit into their business model anyway. With that said, according to Mark Shuttleworth, they are submitting patches upstream to Ubuntu, Apache, and others, thus the avoiding doing any evil. Anyway, this is all speculation, as I don't work there, and there seems to be very little on the subject across the web. So, I'll concede. -- . O . O . O . . O O . . . O . . . O . O O O . O . O O . . O O O O . O . . O O O O . O O O signature.asc Description: OpenPGP digital signature /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 07/21/2010 07:40 AM, Michael Torrie wrote: On 07/20/2010 10:31 PM, Levi Pearson wrote: Except that 'more liberal' literally means more free. :) But but the tea party says that liberal means socialist fascist extremist baby killer! Doesn't sound more free to me! We're all on to your Marxist ways! Don't try to fool us! /me waits for another political thread to erupt -- . O . O . O . . O O . . . O . . . O . O O O . O . O O . . O O O O . O . . O O O O . O O O signature.asc Description: OpenPGP digital signature /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Wed, 2010-07-21 at 00:09 -0600, Levi Pearson wrote: At work we have a wireshark plugin for an in-house protocol that we don't distribute. We also developed a couple of decoders for protocols we're helping to develop through IEEE, and we submitted those back upstream. You can find me in the AUTHORS file, though it may not be in a stable release version yet. Sometimes it's useful to have our in-house protocol and the IEEE draft protocols in the same build, but we've got to be careful not to distribute those builds when we're sending out builds with the draft protocol updates to people we're working with. It's a minor annoyance, but annoying nonetheless. Two things: (1) When you consider how much work has gone into Wireshark, and the fact that you didn't pay a dime for it, whining about managing a plugin is petty. (2) Consider how the situation would be different if you had used a proprietary product instead. You probably still wouldn't choose to distribute your secret sauce plugin. Assuming you could share the tool the same way you are with Wireshark, which you probably couldn't. In other words, you'd be in the same or an even more awkward position. Sounds to me like RMS is annoying ergo you must find any excuse to attack the GPL. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Wed, 2010-07-21 at 07:51 -0600, Aaron Toponce wrote: On 07/21/2010 07:40 AM, Michael Torrie wrote: But but the tea party says that liberal means socialist fascist extremist baby killer! Doesn't sound more free to me! We're all on to your Marxist ways! Don't try to fool us! /me waits for another political thread to erupt Nah, we've all known Torrie for years. Everyone knows he's a Canadian^wMarxist. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: GPL worldview WAS Re: Rchard Stallman vs Darl McBride
IBM did invest in Unix, but under a proprietary license, and called it AIX. Lost of companies have invested in BSD Unix, but compared to Linux, few have contributed back. Richard On Wednesday, July 21, 2010 07:29:53 Michael Torrie torr...@gmail.com wrote: On 07/21/2010 05:18 AM, justin wrote: On Tuesday, July 20, 2010, Michael Torrie torr...@gmail.com wrote: The GPL levels the playing field. IBM's contributions can't be used against them, as they could be with a BSD license. Linux gets better for everyone. I'm not sure I follow. How could IBM's contributions be used against them under a BSD license? Simply put, anything IBM did with a BSD project (and where their code was released under the BSD), could be used in a closed, proprietary way in a product that competes with IBM. Whereas with the GPL, IBM's contributions cannot be closed. If a company chooses to make a superior product based on IBM's contributions under the GPL, IBM is free to incorporate their changes as well (it's only fair after all). Level playing field. IBM works with many open source licenses, but I believe that the GPL is one of the reasons IBM chose (at one time anyway) to invest so heavily in Linux. I do not believe IBM would have invested in the same way in BSD Unix. snip /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: GPL worldview WAS Re: Rchard Stallman vs Darl McBride
On Wed, Jul 21, 2010 at 8:38 AM, Richard Esplin richard-li...@esplins.org wrote: IBM did invest in Unix, but under a proprietary license, and called it AIX. Lost of companies have invested in BSD Unix, but compared to Linux, few have contributed back. All sorts of companies invested in and contributed back to BSD Unix. The early Unix culture was all about sharing and contributing things between people who had the source code, even when it technically violated license agreements. This was before personal computers, though, or at least before you could run Unix on them, so most users worked for companies or universities. USENIX, which was the original Unix Users Group, was where a lot of this happened. Anyway, it's true that Linux has taken over the role of the primary collaborative Unix OS, and that the GPL plays a role in that. It's not the *only* factor involved in that, though. Linux was patterned after SVR4, which was what most of the commercial Unix vendors were based on, and made porting to it easier. Linux developers had a very different philosophy on how to work and what could/should be added to the OS. BSD guys were a lot more conservative about change, while Linux core developers embraced changes eagerly. Linux was born on the PC and naturally attracted PC developers, which was a much larger group than the server/workstation users that were involved in BSD development. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Wed, Jul 21, 2010 at 8:16 AM, Stuart Jansen sjan...@buscaluz.org wrote: On Wed, 2010-07-21 at 00:09 -0600, Levi Pearson wrote: At work we have a wireshark plugin for an in-house protocol that we don't distribute. We also developed a couple of decoders for protocols we're helping to develop through IEEE, and we submitted those back upstream. You can find me in the AUTHORS file, though it may not be in a stable release version yet. Sometimes it's useful to have our in-house protocol and the IEEE draft protocols in the same build, but we've got to be careful not to distribute those builds when we're sending out builds with the draft protocol updates to people we're working with. It's a minor annoyance, but annoying nonetheless. Two things: (1) When you consider how much work has gone into Wireshark, and the fact that you didn't pay a dime for it, whining about managing a plugin is petty. I did say it was a pretty minor annoyance. (2) Consider how the situation would be different if you had used a proprietary product instead. You probably still wouldn't choose to distribute your secret sauce plugin. Assuming you could share the tool the same way you are with Wireshark, which you probably couldn't. Of course we wouldn't distribute the secret sauce plugin. If a plugin API was available (which would be the only way it would be useful to us) we'd also still distribute the plugins for the protocol decoders we're distributing, if it was permissible. The API would undoubtedly make distributing plugins separately from the main package easier than Wireshark does, which would completely alleviate the annoyance, since we wouldn't have to maintain separate builds. There's no motivation for Wireshark to make this easy, since they prefer you to not write them as plugins and to submit them back to the main repository. If there was a proprietary tool that had met our needs in this way and Wireshark wasn't available, we would have paid for it without a second thought. Probably would have been one of the cheaper tools we use. If neither had been available, we'd have written our own tools, which would definitely be more annoying, but licenses would have little to do with that. Sounds to me like RMS is annoying ergo you must find any excuse to attack the GPL. I find RMS *and* the GPL annoying. I find Free Software fanatics even more annoying when they go preaching. However, I think the GPL can be a useful tool, and I might use it or something like it in some circumstances myself. I just think it's given a little bit too much credit sometimes for the software situation today, and I think a lot of people use it out of adherence to Free Software Philosophy rather than for technical reasons. That's their right, but I'm likewise free to share my opinion about it. :) I didn't start my part in this thread with the intention of making you think I was whining about the GPL. My original point is that I didn't want to hear a presentation from RMS, because I didn't think it would contain useful technical information. Other people drove the discussion towards GPL and its importance; my comments on it have not been intended to attack it, but to provide a perspective on it from outside the FSF worldview. I think I've been pretty fair (if sometimes not terribly specific, opening myself to clarifying missives from people who assumed I didn't understand it) in my assertions about the GPL, and I've admitted that I might be wrong about its level of influence on the software landscape today--it's definitely debatable. My examples were not meant to whine about GPL, but to illustrate that it hasn't had much of an effect on companies I've been involved with contributing back to projects. The things we kept back would have been kept back under a more permissive license, and the things we contributed would have been contributed as well, because it just made sense to do so. The GPL was mostly irrelevant, aside from creating a minor annoyance. This is purely anecdotal, but so have been the other examples I've seen. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
I want to respond to two different, related posts. On 07/14/2010 04:21 PM, Levi Pearson wrote: snip Seriously guys, this is a USERS group. Neither McBride nor Stallman would have anything useful to say about USING Linux. One of them is out to make a buck, and happened to end up on the wrong side of a lawsuit. The other is the leader of a social/political movement, and seems to be only really interested in advocating that movement. snip One of the roles of an open source software users group is to educate users on the rights open source gives them, thereby helping them to value those rights. An open license has value that can be compelling even when the software is not necessarily the best technical fit. At my current job, our marketing team recently surveyed our customers to determine why they selected our open source software over proprietary alternatives. More than half of our customers listed the open source license as a major reason they selected our product. Many customers reported that the license played a bigger role than the technical fit of the product. Licenses matter, especially to an open source users group. On Saturday, July 17, 2010 15:56:31 Levi Pearson levipear...@gmail.com wrote: snip What's the big deal about 'getting hijacked by powerful corporate interests', anyway? It was clearly possible to write significant software to be released under a BSD license, because BSD did it. Certainly lots of noise and many heated arguments were created by the GPL, but I'm not convinced that it was a major factor in getting stuff written. snip I have often been involved in deciding what license a business should use for a new software project. There are trade-offs to each license, and I like Bruce Peren's advice here: http://itmanagement.earthweb.com/osrc/article.php/3803101/Bruce-Perens-How-Many-Open-Source-Licenses-Do-You-Need.htm The GPL provides some strong protections for code creators. * Reciprocity plays to a sense of justice I am not going to use my free time working on a project which will primarily enrich Bill Gates and his share-holders (or any other company). * Reciprocity provides the hope of additional contributions When I select the GPL, I am optimistic that someone else will like my code enough to contribute to it, and respect the license enough that I will benefit from their additions. The BSD provides a warm- fuzzy feeling that my code might help someone else, but I have no legal expectation that I will benefit in return. * Share-alike creates more open code The legal obligation of reciprocity can act as a multiplier on hobby code contributions, because corporations that want to improve the code have additional reasons to persuade management to release their improvements. At two different companies, I have been able to release improvements to open source code because the project we wanted to use was licensed under the GPL. If it had been BSD licensed, that code would be rotting on some small team within the companies. The programmers who produced that code were not paid to create it, but they benefit from my team being paid to improve it. * Share-alike protects the programmer When I am paid to learn and contribute to GPL code licensed code, I know that the time and effort I have invested will be transferable to other customers and employers. Other open source licenses have much weaker protections. The GPL might not be the right fit for every line of code you create, but software licenses are important and ignoring the license on code you create or use results in giving up rights that you might prefer to retain or that you should retain on behalf of your company. Many technical people do not appreciate the role that intellectual property rights play in our society or how the rules surrounding intellectual property factor into business decision making. Technical people are often so focused on the technical quality of an engineering contribution that they do not grasp how that contribution is, or is not, benefiting a business, a profession, or society. Lawrence Lessig's writings on these matters is very illuminating. _Code_ is a good place to start: http://codev2.cc/download+remix/ Richard Esplin /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 5:02 PM, Richard Esplin richard-li...@esplins.org wrote: I want to respond to two different, related posts. On 07/14/2010 04:21 PM, Levi Pearson wrote: snip Seriously guys, this is a USERS group. Neither McBride nor Stallman would have anything useful to say about USING Linux. One of them is out to make a buck, and happened to end up on the wrong side of a lawsuit. The other is the leader of a social/political movement, and seems to be only really interested in advocating that movement. snip One of the roles of an open source software users group is to educate users on the rights open source gives them, thereby helping them to value those rights. An open license has value that can be compelling even when the software is not necessarily the best technical fit. At my current job, our marketing team recently surveyed our customers to determine why they selected our open source software over proprietary alternatives. More than half of our customers listed the open source license as a major reason they selected our product. Many customers reported that the license played a bigger role than the technical fit of the product. Licenses matter, especially to an open source users group. Yes, they do matter, and information about how to select an appropriate license for your needs is actually a valid technical topic. Licenses are legal tools to specific ends, and knowing how to use those tools and what their usage implies is important whether you subscribe to a particular social view or not. However, advocating social views, which is RMS's modus operandi these days, is not the same thing as giving technical instruction/advice on selecting a license. Yes, there's an argument that the GPL is an important license now and thus getting a treatise on the philosophy behind it and a call to follow that philosophy is appropriate, and I'll have to concede that it's a somewhat reasonable one that might appeal to a good portion of the group. I just don't think it's very useful, myself. I'd rather hear a less biased overview. On Saturday, July 17, 2010 15:56:31 Levi Pearson levipear...@gmail.com wrote: snip What's the big deal about 'getting hijacked by powerful corporate interests', anyway? It was clearly possible to write significant software to be released under a BSD license, because BSD did it. Certainly lots of noise and many heated arguments were created by the GPL, but I'm not convinced that it was a major factor in getting stuff written. snip I have often been involved in deciding what license a business should use for a new software project. There are trade-offs to each license, and I like Bruce Peren's advice here: http://itmanagement.earthweb.com/osrc/article.php/3803101/Bruce-Perens-How-Many-Open-Source-Licenses-Do-You-Need.htm The GPL provides some strong protections for code creators. * Reciprocity plays to a sense of justice I am not going to use my free time working on a project which will primarily enrich Bill Gates and his share-holders (or any other company). If you're using your free time to work on a project, presumably you're primarily enriching yourself. If it's not personally enriching, you probably shouldn't be doing it in your free time! If you don't want someone else to make money from your efforts, either don't release it, or release it with a no-commercial-use license. The GPL doesn't prohibit selling software, it just obliges people who distribute modified versions to make source available. If you're worried about them taking credit for your work, there are licenses that require that credit/attribution be given to you. In other words, I don't see how the above description of motivations is related to a sense of justice as much as it is related to compelling other people to do what you want them to do with their efforts. * Reciprocity provides the hope of additional contributions When I select the GPL, I am optimistic that someone else will likeIf my code enough to contribute to it, and respect the license enough that I will benefit from their additions. The BSD provides a warm- fuzzy feeling that my code might help someone else, but I have no legal expectation that I will benefit in return. GPL provides no expectation of benefit, only an expectation of access to source code of changes. There's no reason that the GPL would compel someone to make changes that would be useful to you or in a form you would like. Cooperation can't be compelled by license, and it makes business sense to cooperate outside of license compulsion. I don't think the GPL was required to bring about this understanding, but that's certainly debatable. * Share-alike creates more open code The legal obligation of reciprocity can act as a multiplier on hobby code contributions, because corporations that want to improve the code have additional reasons to persuade management to
Re: Rchard Stallman vs Darl McBride
On Tue, 20 Jul 2010, Levi Pearson wrote: Giving up the right to not distribute parts of your source code is a pretty significant right to give up for some pretty dubious 'protection' offered by the GPL. It's not clear to me what you're saying here. What about the GPL would cause one to give up the right to not distribute parts of your source code? Jon -- Jon Jensen End Point Corporation http://www.endpoint.com/ /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 7/20/2010 7:23 PM, Jon Jensen wrote: On Tue, 20 Jul 2010, Levi Pearson wrote: Giving up the right to not distribute parts of your source code is a pretty significant right to give up for some pretty dubious 'protection' offered by the GPL. It's not clear to me what you're saying here. What about the GPL would cause one to give up the right to not distribute parts of your source code? Jon You can't release part of a project under GPL, and if you do release it under GPL, you give up the right to keep some of your code to yourself. At least, I believe that's what he's saying. -Tod Hansmann /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
GPL worldview WAS Re: Rchard Stallman vs Darl McBride
Responses are inline. On Tuesday, July 20, 2010 19:14:43 Levi Pearson levipear...@gmail.com wrote: snip The GPL provides some strong protections for code creators. * Reciprocity plays to a sense of justice I am not going to use my free time working on a project which will primarily enrich Bill Gates and his share-holders (or any other company). If you're using your free time to work on a project, presumably you're primarily enriching yourself. If it's not personally enriching, you probably shouldn't be doing it in your free time! If you don't want someone else to make money from your efforts, either don't release it, or release it with a no-commercial-use license. The GPL doesn't prohibit selling software, it just obliges people who distribute modified versions to make source available. If you're worried about them taking credit for your work, there are licenses that require that credit/attribution be given to you. In other words, I don't see how the above description of motivations is related to a sense of justice as much as it is related to compelling other people to do what you want them to do with their efforts. Personal enrichment is only one motivation for contributing to a software project. From the perspective of what is best for the open source ecosystem, it is perhaps one of the less valuable motivations. You are miss-characterizing the primary motivation. For people like me, the point is to maximize social good. The worry is not commercial use or miss-attribution. The point is to build something that would have economic value, and give it away for free because it will benefit others. Asking those who want to benefit from my work to contribute back their changes seems like a just request. Letting others leverage the economic value of my contribution without giving anything in return appears unjust. * Reciprocity provides the hope of additional contributions When I select the GPL, I am optimistic that someone else will likeIf my code enough to contribute to it, and respect the license enough that I will benefit from their additions. The BSD provides a warm- fuzzy feeling that my code might help someone else, but I have no legal expectation that I will benefit in return. GPL provides no expectation of benefit, only an expectation of access to source code of changes. There's no reason that the GPL would compel someone to make changes that would be useful to you or in a form you would like. Cooperation can't be compelled by license, and it makes business sense to cooperate outside of license compulsion. I don't think the GPL was required to bring about this understanding, but that's certainly debatable. In legal terms, having an expectation of access to source code of derived works is a benefit with economic value. Though not preferable, cooperation can be compelled by license. Look at the Linksys WRT54G experience. It is true that it is preferable to cooperate outside of license compulsion, but the GPL encourages such cooperation by creating a level playing field with clearly defined legal expectations. This allows competitors to cooperate while avoiding the free-rider problem. In a competitive free-market economy, it is too risky to cooperate without some legal agreement. The Apache Project and the Eclipse Foundation provide the legal basis of cooperation through the legal agreements of their governing boards. I personally think the software license is a lower bandwidth way of defining the terms of the cooperation. * Share-alike creates more open code The legal obligation of reciprocity can act as a multiplier on hobby code contributions, because corporations that want to improve the code have additional reasons to persuade management to release their improvements. At two different companies, I have been able to release improvements to open source code because the project we wanted to use was licensed under the GPL. If it had been BSD licensed, that code would be rotting on some small team within the companies. The programmers who produced that code were not paid to create it, but they benefit from my team being paid to improve it. I have worked with companies that were extremely hesitant to touch anything GPL-related due to worries about license contamination of important IP. They would not release source related to that IP under a free license under any circumstance, though they might find it useful to cooperate with others on non-core software. GPL would prevent that non-core cooperation due to the worry of being forced to release core IP. The open source projects I have contributed to are things that we would have contributed to regardless of license, because cooperation was one of the goals from the beginning. Companies who do not want to disclose their improvements to GPL code should not be using that code.
Re: Rchard Stallman vs Darl McBride
On Tue, 20 Jul 2010, Tod Hansmann wrote: You can't release part of a project under GPL, and if you do release it under GPL, you give up the right to keep some of your code to yourself. At least, I believe that's what he's saying. There is nothing in the GPL that says you must distribute any code you write that is derivative of GPL'd code, if you don't distribute it. If you distribute code derivative of GPL'd code, then it must be licensed under the GPL. But you don't have to distribute it. For example, a company can make internal modifications to Linux and is under no obligation to distribute them or tell anyone anywhere about it. I believe internal copying within a company is not considered distribution under copyright law due to the idea of corporate personhood. However, if they do distribute the code to others, whether in exchange for money or not, it must be licensed under the GPL to comply with the GPL license of the code they're building upon. Otherwise, they have no license to use that copyrighted original code. One option that occasionally is used: You can get permission from the original copyright holder to use their code under another license than the GPL, maybe for free because they like you, or in exchange for money or other valuables. This is all just a reflection of the fact that under current copyright law, all creative works are presumed copyrighted by their creator by default, and the only thing that makes it legal for another party to copy the copyrighted work (except for fair use etc.) is permission, whether through a stock license like the GPL, or personal permission. Jon -- Jon Jensen End Point Corporation http://www.endpoint.com/ /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 7:32 PM, Tod Hansmann t...@todandlorna.com wrote: On 7/20/2010 7:23 PM, Jon Jensen wrote: On Tue, 20 Jul 2010, Levi Pearson wrote: Giving up the right to not distribute parts of your source code is a pretty significant right to give up for some pretty dubious 'protection' offered by the GPL. It's not clear to me what you're saying here. What about the GPL would cause one to give up the right to not distribute parts of your source code? Jon You can't release part of a project under GPL, and if you do release it under GPL, you give up the right to keep some of your code to yourself. At least, I believe that's what he's saying. Yes, that's it. If there's a chunk of your code that's linked to the rest that you want to keep private, for whatever reason, you cannot use the GPL with the rest of your code. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, 2010-07-20 at 20:35 -0600, Levi Pearson wrote: Yes, that's it. If there's a chunk of your code that's linked to the rest that you want to keep private, for whatever reason, you cannot use the GPL with the rest of your code. Not entirely true. If you control the copyright on all of the code, you can do whatever you want. That's what makes dual licensing possible, and one of the reason some companies insist on copyright assignment before accepting contributions. Of course, if you depend on someone else's GPL'd code, you don't control the copyright of all the code and are therefore subject to the GPL. You'd get PR flack for releasing a crippled project, but you can do it. The GPL can't dictate terms to the copyright holder. -- Stuart Jansen sjan...@buscaluz.org /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 8:34 PM, Jon Jensen j...@endpoint.com wrote: On Tue, 20 Jul 2010, Tod Hansmann wrote: You can't release part of a project under GPL, and if you do release it under GPL, you give up the right to keep some of your code to yourself. At least, I believe that's what he's saying. There is nothing in the GPL that says you must distribute any code you write that is derivative of GPL'd code, if you don't distribute it. If you distribute code derivative of GPL'd code, then it must be licensed under the GPL. But you don't have to distribute it. This is true, though you might make the GPL software's authors mad if they find out about it. I was assuming that we were talking about code that was to be distributed, since otherwise you don't have to license your code at all. This is all just a reflection of the fact that under current copyright law, all creative works are presumed copyrighted by their creator by default, and the only thing that makes it legal for another party to copy the copyrighted work (except for fair use etc.) is permission, whether through a stock license like the GPL, or personal permission. Yes, but you can license your software to grant permission to use, copy, modify, and redistribute it without preaching to people about rights and freedoms and whatnot, and without forcing them to open any software they might link with yours. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
GPL does not tie your hands necessarily - Re: Rchard Stallman vs Darl McBride
On 07/20/2010 08:35 PM, Levi Pearson wrote: Yes, that's it. If there's a chunk of your code that's linked to the rest that you want to keep private, for whatever reason, you cannot use the GPL with the rest of your code. It pains me to see misunderstanding of the GPL and how it works. If your wrote *all* the code yourself, you absolutely can release only part of it under the GPL. Most projects tend to have a written exception clause to the GPL, but that is not strictly required. It's your code; you can license it anyway you want. So I could write a program with a closed-source plugin, and release everything but the plugin under the GPL. Any derivatives of my code must be GPL, since that's the license I specified. My having a closed module in no way changes that. Derivatives of your program (the GPL'd part) could or could not ship a closed module, depending on exceptions that you as the copyright holder make and append to your license. See the Linux Kernel's license, or GCC's license for exceptions to the GPL that are allowed for. On the other hand, if I made a derivative of some GPL's project and then wanted to also ship a closed module for use with that, then you obviously cannot, because that would violate the terms that someone else (who owns the copyright on the code) dictated. You could, of course, get the copyright owner to grant an exception for your closed module. In this case, if the GPL doesn't work for you, write your own code, negotiate a different license with the copyright holder, or find some other code whose license works for you. I hope that makes sense. Copyright and licenses are not magical. Nothing is *automatic*. Certainly in the case of the GPL, the license is there to ensure the freedom of the *developer* first. The end user is of course always free, provided he does not distribute, as the GPL does not require the end user to agree to anything to use GPL'd software. Just try to use Java on a nuclear submarine... /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, 2010-07-20 at 20:50 -0600, Levi Pearson wrote: Yes, but you can license your software to grant permission to use, copy, modify, and redistribute it without preaching to people about rights and freedoms and whatnot, and without forcing them to open any software they might link with yours. Clearly you need a short, overweight, bearded, unwashed man to visit you at 5am and discuss software licensing in depth. My co-workers probably won't be happy to discover I've decided to stop showering, but... we all have to make sacrifices. See you in a week. -- Stuart Jansen sjan...@buscaluz.org P.S. Will you supply the donuts or should I? /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 8:49 PM, Stuart Jansen sjan...@buscaluz.org wrote: On Tue, 2010-07-20 at 20:35 -0600, Levi Pearson wrote: Yes, that's it. If there's a chunk of your code that's linked to the rest that you want to keep private, for whatever reason, you cannot use the GPL with the rest of your code. Not entirely true. If you control the copyright on all of the code, you can do whatever you want. That's what makes dual licensing possible, and one of the reason some companies insist on copyright assignment before accepting contributions. Of course, if you depend on someone else's GPL'd code, you don't control the copyright of all the code and are therefore subject to the GPL. That's a good point, and a valid way to work around the problem if you don't want/need to take advantage of any GPL libraries you don't own the copyright to in the non-private portions. The GPL just creates what seems sometimes like a minefield for those who want to keep some things private. That's the intention of it, really. I think it makes more sense in this case to just use a permissive license. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, 2010-07-20 at 21:11 -0600, Levi Pearson wrote: The GPL just creates what seems sometimes like a minefield for those who want to keep some things private. That's the intention of it, really. I think it makes more sense in this case to just use a permissive license. Dude, that's like arguing: The DMZ is preventing the reintegration of Korea, we should just let the North invade the South and the conflict will be ended. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 07/20/2010 09:11 PM, Levi Pearson wrote: The GPL just creates what seems sometimes like a minefield for those who want to keep some things private. That's the intention of it, really. I think it makes more sense in this case to just use a permissive license. *Any* third-party code license can lead to a minefield for developers. In short, if you don't know how each and every pieces of third party code you are using is licensed and how it affects how you intend to license your code, you are in trouble. This is true in the regular, proprietary world, or in the open source world. Course if everyone used the GPL there'd be no problem. :) /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: GPL worldview WAS Re: Rchard Stallman vs Darl McBride
On 07/20/2010 08:21 PM, Richard Esplin wrote: Personal enrichment is only one motivation for contributing to a software project. From the perspective of what is best for the open source ecosystem, it is perhaps one of the less valuable motivations. In fact, IBM contributes to GPL'd software precisely for *commercial* reasons. In IBM's mind, contributing to GPLd software like the Linux kernel helps them make money by having a stronger, more viable platform. The GPL levels the playing field. IBM's contributions can't be used against them, as they could be with a BSD license. Linux gets better for everyone. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, 20 Jul 2010, Levi Pearson wrote: Yes, but you can license your software to grant permission to use, copy, modify, and redistribute it without preaching to people about rights and freedoms and whatnot, and without forcing them to open any software they might link with yours. Yes, that's certainly an option. We should just remember that everyone whether they go out of their way to assert it or not denies the freedom to use, copy, modify, and redistribute anything they write unless they go out of their way to grant those permissions to others. So the GPL may seem draconian, but if you had to specifically claim all the rights that copyright law gives you, you'd sound pretty preachy and fussy about rights and whatnot. I think in light of the new default everything is copyrighted by default legal situation (only since 1976 in the U.S.) the GPL is still pretty friendly. In other words, I don't see the more liberal licenses such as BSD or MIT being on any higher moral ground, or freer, than the GPL. They just have different tradeoffs. Jon -- Jon Jensen End Point Corporation http://www.endpoint.com/ /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, 2010-07-20 at 21:35 -0600, Michael Torrie wrote: Course if everyone used the GPL there'd be no problem. :) By that logic, the DBADL is also a viable option. http://philsturgeon.co.uk/code/dbad-license /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: GPL does not tie your hands necessarily - Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 8:53 PM, Michael Torrie torr...@gmail.com wrote: On 07/20/2010 08:35 PM, Levi Pearson wrote: Yes, that's it. If there's a chunk of your code that's linked to the rest that you want to keep private, for whatever reason, you cannot use the GPL with the rest of your code. It pains me to see misunderstanding of the GPL and how it works. It was not so much a misunderstanding as an oversimplification with hidden assumptions, but your clarification is good information. If your wrote *all* the code yourself, you absolutely can release only part of it under the GPL. Yes, but that negates the power of using open source building blocks. I assumed some other copyright holders in the mix, either from the beginning or from accepting contributions under the GPL without copyright reassignment. Most projects tend to have a written exception clause to the GPL, but that is not strictly required. It's your code; you can license it anyway you want. So I could write a program with a closed-source plugin, and release everything but the plugin under the GPL. Any derivatives of my code must be GPL, since that's the license I specified. My having a closed module in no way changes that. Derivatives of your program (the GPL'd part) could or could not ship a closed module, depending on exceptions that you as the copyright holder make and append to your license. See the Linux Kernel's license, or GCC's license for exceptions to the GPL that are allowed for. Yes, but that's GPL-with-exceptions, not vanilla GPL. Structuring things correctly and coming up with the necessary exceptions and all that seem needlessly complicated to me. On the other hand, if I made a derivative of some GPL's project and then wanted to also ship a closed module for use with that, then you obviously cannot, because that would violate the terms that someone else (who owns the copyright on the code) dictated. You could, of course, get the copyright owner to grant an exception for your closed module. In this case, if the GPL doesn't work for you, write your own code, negotiate a different license with the copyright holder, or find some other code whose license works for you. I don't think most sizeable GPL works have a single copyright holder, unless they were set up in advance with the intention of providing alternate licenses as well. This makes a vast amount of otherwise useful software not very useful at all in some situations, even if you're willing to pay. I hope that makes sense. Copyright and licenses are not magical. Nothing is *automatic*. Certainly in the case of the GPL, the license is there to ensure the freedom of the *developer* first. The end user is of course always free, provided he does not distribute, as the GPL does not require the end user to agree to anything to use GPL'd software. Just try to use Java on a nuclear submarine... It doesn't seem to me that the GPL is about ensuring the freedom of the developer, unless of course you mean contributors other than the initial author. Indeed, the preamble states that the purpose is to allow people who receive the program to access the source code, make modifications, and re-distribute it. The initial author already has the freedom to do these things. You could say it grants the freedom to not have the code incorporated into a non-free system, but that's a pretty odd way to frame what it does. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 9:41 PM, Jon Jensen j...@endpoint.com wrote: On Tue, 20 Jul 2010, Levi Pearson wrote: So the GPL may seem draconian, but if you had to specifically claim all the rights that copyright law gives you, you'd sound pretty preachy and fussy about rights and whatnot. I think in light of the new default everything is copyrighted by default legal situation (only since 1976 in the U.S.) the GPL is still pretty friendly. In other words, I don't see the more liberal licenses such as BSD or MIT being on any higher moral ground, or freer, than the GPL. They just have different tradeoffs. Except that 'more liberal' literally means more free. :) I get your point, though, and that's what I said at the beginning of this. I don't care for the social movement surrounding the GPL, and I would rather attend a presentation on technical topics than on Free Software evangelism. As a document granting specific rights, the GPL is a useful tool in some circumstances. As a literary work, it reads like the love child of a missionary tract and a regular software license. I don't generally like its tradeoffs or its style, though I will concede that it has served a useful purpose. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 7/20/2010 9:42 PM, Stuart Jansen wrote: On Tue, 2010-07-20 at 21:35 -0600, Michael Torrie wrote: Course if everyone used the GPL there'd be no problem. :) By that logic, the DBADL is also a viable option. http://philsturgeon.co.uk/code/dbad-license I wrote this license once, and have some code under it. I like it better than the GPL. YMMV (It's essentially the WTF license, but more professional, no swearing). It works for art and other content as well! -Tod Hansmann Any entity is permitted to copy, redistribute, or modify the accompanying work for any reason whatsoever with no conditions except the condition below and under any new license. Do what you want, but don't blame me. THIS WORK IS PROVIDED BY THE AUTHOR(S) 'AS IS' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS WORK, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, 20 Jul 2010, Levi Pearson wrote: In other words, I don't see the more liberal licenses such as BSD or MIT being on any higher moral ground, or freer, than the GPL. They just have different tradeoffs. Except that 'more liberal' literally means more free. :) Yes, the BSD and MIT licenses are more liberal about what they allow the end user to do. But the GPL is counterintuitively more liberal about what it gives the community, using copyright law in a nontraditional way. If the law itself were more liberal, there'd be no need for either. I get your point, though, and that's what I said at the beginning of this. I don't care for the social movement surrounding the GPL, and I would rather attend a presentation on technical topics than on Free Software evangelism. Sure. Point taken. I think many of us see value in discussing the freedom aspect of the free software movement, and consider it irresponsible not to expose newcomers to the community about it. Many of us have been on this list for a very long time, and to us the topics are well-worn and mostly understood (though not entirely, as tonight's discussion shows). But to newcomers, they deserve to hear it from the FSF perspective as well as the Open Source crowd and wherever else. If you already know that, of course it makes sense to have less interest. The Business Software Alliance, the MPAA, the RIAA, and others make sure the proprietary mindset is widely preached, so I see no need to give them an additional venue to spread their doctrine. (In which class I presume some of the people who don't want to hear from Darl McBride are putting him.) Jon -- Jon Jensen End Point Corporation http://www.endpoint.com/ /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 07/20/2010 08:34 PM, Jon Jensen wrote: There is nothing in the GPL that says you must distribute any code you write that is derivative of GPL'd code, if you don't distribute it. If you distribute code derivative of GPL'd code, then it must be licensed under the GPL. But you don't have to distribute it. For example, a company can make internal modifications to Linux and is under no obligation to distribute them or tell anyone anywhere about it. I believe internal copying within a company is not considered distribution under copyright law due to the idea of corporate personhood. I'm surprised this hasn't been brought up, so I'll mention it. Google keeps an internal proprietary derivative of Ubuntu on their employee workstations. They have never released this Goobuntu to the wild, and as such are under no obligation to release their changes. Even if the code is GPL. http://en.wikipedia.org/wiki/Goobuntu -- . O . O . O . . O O . . . O . . . O . O O O . O . O O . . O O O O . O . . O O O O . O O O signature.asc Description: OpenPGP digital signature /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Tue, Jul 20, 2010 at 10:54 PM, Jon Jensen j...@endpoint.com wrote: On Tue, 20 Jul 2010, Levi Pearson wrote: If the law itself were more liberal, there'd be no need for either. I'm not sure how you could end up with the sort of requirements the GPL has without explicitly stating them in a license. If there's no implicit copyright to creative works, then if you publish them, anyone can modify them and not share the modifications. I guess if the law required all software to have source freely available to those it is distributed to and did not provide copyright provisions for software or allow licenses , that would do it, but that seems awfully unlikely and I'm not sure I'd call it 'more liberal'. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Sat, Jul 17, 2010 at 3:33 PM, Stuart Jansen sjan...@buscaluz.org wrote: Lucky timing might have given Linux a boost, but it doesn't explain why none of the BSDs could retake the lead despite having arguably better technology at times. Linux leads OS X in terms of market share or installed base? By what degree, and when did this happen? /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Alternate History (was Re: Rchard Stallman vs Darl McBride)
Stuart Jansen sjan...@buscaluz.org writes: On Fri, 2010-07-16 at 11:43 -0600, Roger Brown wrote: I agree, if it wasn't for the ATT lawsuit we would probably all be running some BSD flavor today. *sigh* I'm tired of this meme. It should have died years ago. Attributing the success of the Linux kernel to lucky timing is myopic. So, I recently saw an interview with Linus from '93 or so, in which he said that if 386BSD had been released a year or two earlier, he probably wouldn't have created Linux at all. That would, however, have left RMS and the GNU crew still looking for a kernel. I wonder how that would have played out? Maybe the HURD guys would have had some serious pressure to actually get a system working, and we'd have ended up with BSD and GNU HURD as free UNIX-style OS alternatives. Would the GPL have fostered the same community without Linus and his kernel development style? Would BSD-derivatives have become what Linux is today, or would HURD have grown to the same stature? One interesting factoid is that the initial HURD architect said that their initial plan was to base their kernel on the 4.4BSD-Lite kernel. In that case (and this may have been early enough to preempt Linus from writing Linux, too) we really would all be running at least a descendant of a BSD system, even if it may not have been easily recognizable as one. RMS screwed that one up, though, by making the call to go with Mach instead, over the objections of said architect. Anyway, although the success of the Linux kernel can't be solely attributed to lucky timing, the fact that it came into existence at all can be. There were a number of other options poised to take off, and had Linux never arrived, at least some of us would probably be running one of them now. What the software landscape would look like in that case is pretty hard to predict, though. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Mon, 2010-07-19 at 00:28 -0600, Chris wrote: On Sat, Jul 17, 2010 at 3:33 PM, Stuart Jansen sjan...@buscaluz.org wrote: Lucky timing might have given Linux a boost, but it doesn't explain why none of the BSDs could retake the lead despite having arguably better technology at times. Linux leads OS X in terms of market share or installed base? By what degree, and when did this happen? Oh come off it, Mac OS X is hardly a BSD. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Fri, 2010-07-16 at 11:43 -0600, Roger Brown wrote: I agree, if it wasn't for the ATT lawsuit we would probably all be running some BSD flavor today. *sigh* I'm tired of this meme. It should have died years ago. Attributing the success of the Linux kernel to lucky timing is myopic. It ignores the importance of the GPL in encouraging contributions during a time when many people didn't userstand F/OSS development or believe that it was possible to write significant software under a BSD license without it getting hijacked by powerful corporate interests. It ignore the disadvantages to BSD's more heavy weight and bureaucratic approach to coordination. Perhaps Linus Torvald's greatest contribution is not Linux or git but demonstrating that large software projects can be loosely coordinated, innovate quickly, and still maintain an impressive quality level. (In fact, it's starting to look like the Linux development model is the only viable model for very large projects to achieve success. Remember the train wreck that was Vista? Microsoft attributed much of their failure to too much project coordination overhead. There response was increased decentralization, producing marked improvement in Windows 7.) Let's not forget that Linux has gone through periods of significant growing pain. Anyone remember the shameful early days of the 2.4 kernel? If ever there were a time for one of the BSDs to step up and replace Linux, that was it. Lucky timing might have given Linux a boost, but it doesn't explain why none of the BSDs could retake the lead despite having arguably better technology at times. -- Stuart Jansen sjan...@buscaluz.org /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Fri, 2010-07-16 at 16:03 -0600, Shane Hathaway wrote: My experience has been different. More than a decade ago, Eric Raymond's papers convinced me to try something beyond Windows/Mac and take a serious look at Linux. I had to be convinced that free software and the people behind it were worth my time. Raymond's perspective made sense to me; Stallman's did not. So I think Raymond's writings were very valuable for inviting new people like me into the existing hacker culture. Consider yourself lucky that you were already part of that culture. Both ESR and RMS played important parts in informing my software politics. But it bugs me that both act like they represent me and everyone else in the community (whatever that means). I hope I don't sound too conceited when I say I've outgrown them. Both have played valuable roles, but their propaganda is too narrow. It hints and reality, but does not describe it. Bottom line: I respect both of them, but ESR does not represent me and neither does RMS. -- Stuart Jansen sjan...@buscaluz.org /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Sat, Jul 17, 2010 at 3:33 PM, Stuart Jansen sjan...@buscaluz.org wrote: On Fri, 2010-07-16 at 11:43 -0600, Roger Brown wrote: I agree, if it wasn't for the ATT lawsuit we would probably all be running some BSD flavor today. *sigh* I'm tired of this meme. It should have died years ago. Attributing the success of the Linux kernel to lucky timing is myopic. It ignores the importance of the GPL in encouraging contributions during a time when many people didn't userstand F/OSS development or believe that it was possible to write significant software under a BSD license without it getting hijacked by powerful corporate interests. What's the big deal about 'getting hijacked by powerful corporate interests', anyway? It was clearly possible to write significant software to be released under a BSD license, because BSD did it. Certainly lots of noise and many heated arguments were created by the GPL, but I'm not convinced that it was a major factor in getting stuff written. It ignore the disadvantages to BSD's more heavy weight and bureaucratic approach to coordination. Perhaps Linus Torvald's greatest contribution is not Linux or git but demonstrating that large software projects can be loosely coordinated, innovate quickly, and still maintain an impressive quality level. I'll have to agree with you that the development model for Linux ended up being a major source of its success, but I don't think it has anything to do with the license. I can easily imagine an alternate future where GNU didn't exist, Linux was BSD licensed, and the userland tools were all forks or rewrites of BSD tools. I think for the majority of people involved, the point was to have a working, hackable, no-cost UNIX system. The details of licenses were a sideline issue brought to the forefront by RMS and the BSD lawsuit. (In fact, it's starting to look like the Linux development model is the only viable model for very large projects to achieve success. Remember the train wreck that was Vista? Microsoft attributed much of their failure to too much project coordination overhead. There response was increased decentralization, producing marked improvement in Windows 7.) It also helped that Windows 7 was largely a refinement of Vista, rather than another major rewrite. There are some serious downsides to the Linux development model, too, but there's no doubt it's been successful. Let's not forget that Linux has gone through periods of significant growing pain. Anyone remember the shameful early days of the 2.4 kernel? If ever there were a time for one of the BSDs to step up and replace Linux, that was it. Lucky timing might have given Linux a boost, but it doesn't explain why none of the BSDs could retake the lead despite having arguably better technology at times. Linux and GNU would pretty much have to have been absent for BSD to take over, but having 'been there', so to speak, I know that the time was ripe for a free UNIX system to spread to home hobbyists. PCs were finally powerful enough to run them, people wanted to run them at home, and free UNIX systems existed. It was just time for it to happen. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Sat, 2010-07-17 at 15:56 -0600, Levi Pearson wrote: What's the big deal about 'getting hijacked by powerful corporate interests', anyway? Psychology. While some people hack purely for the love of hacking, others depend on the terms of the GPL to ensure reciprocity. Especially fifteen years ago when their was less social pressure to play nice with the F/OSS community. http://en.wikipedia.org/wiki/Reciprocity_%28social_psychology%29 /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
Hi, haven't read the list email too much -- but would like to come to a meeting sometime. I just wanted to chime in with an amen on this comment. :) On Thu, Jul 15, 2010 at 10:04 AM, Levi Pearson levipear...@gmail.com wrote: ... The only reason BSD didn't take the popularity surge that Linux did was a small and short-lived licensing issue that existed right at the cusp of things. I agree, if it wasn't for the ATT lawsuit we would probably all be running some BSD flavor today. I started out with the original iServer using BSD/OS and FreeBSD back in the day. I remember when Microsoft first bought Hotmail.com how for a couple of years they ran it on FreeBSD because they couldn't get it to run on Windows. :) Yahoo used to be all FreeBSD but now it looks like only my.yahoo.com is using it. (http://searchdns.netcraft.com/?host=yahoo.com ) And Microsoft at some point in the 90's copied code from FreeBSD for their TCP/IP stack since the Microsoft one was sucky. Anyway, I wish that more people were aware of or remembered to give proper credit to the BSD core upon which the Internet was built. FreeBSD used to power practically all of the high traffic websites back in the 90's. But I know this is a Linux Group and Ubuntu Linux is now my main desktop os. :) Just trying to give credit where credit is due. Thanks to those that maintain this list. Lots of good discussion, Roger Brown /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Fri, 16 Jul 2010 11:43:57 -0600 Roger Brown downtownrogbr...@gmail.com wrote: But I know this is a Linux Group and Ubuntu Linux is now my main desktop os. :) Just trying to give credit where credit is due. As indeed we should. One book that gives some of the background is Eric Steven Raymond, The Art of Unix Programming, 2003, http://www.faqs.org/docs/artu/. It's ESR's usual iconoclastic (pun intended) writing, but well worth it for programmers and those who have the conceit that they can manage programmers. M$ used BSD code for their W9x programs as well, including ftp.exe and route.exe. You can verify this by running strings against them and reading the copyright notices. -- Charles Curley /\ASCII Ribbon Campaign Looking for fine software \ /Respect for open standards and/or writing? X No HTML/RTF in email http://www.charlescurley.com/ \No M$ Word docs in email Key fingerprint = CE5C 6645 A45A 64E4 94C0 809C FFF6 4C48 4ECD DFDB /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Fri, Jul 16, 2010 at 12:16 PM, Charles Curley charlescur...@charlescurley.com wrote: On Fri, 16 Jul 2010 11:43:57 -0600 Roger Brown downtownrogbr...@gmail.com wrote: But I know this is a Linux Group and Ubuntu Linux is now my main desktop os. :) Just trying to give credit where credit is due. As indeed we should. One book that gives some of the background is Eric Steven Raymond, The Art of Unix Programming, 2003, http://www.faqs.org/docs/artu/. It's ESR's usual iconoclastic (pun intended) writing, but well worth it for programmers and those who have the conceit that they can manage programmers. ESR annoys me almost as much as RMS. At least RMS was a real hacker; ESR is just kind of a hacker groupie with an inflated sense of self-importance. He's had some good ideas, he's made a few code contributions, and he's definitely been heavily involved in the periphery of the whole open source/free software movement, but my point is that the 'movement' aspect of software sharing and group development is useless and annoying. People naturally share ideas and help each other to work on projects. The early history of computer software 'hacker' culture shows this, and it worked pretty well without huge egos, figureheads, and organized movements. The fact that 'users groups' and the like have existed almost as long as computers have shows this, too. It worked just fine before RMS and ESR went on their crusades, and I believe we would have largely ended up in the same place we are now without the non-technical contributions of either of them. If you want my idea of a hacker hero, take a look at Guy Steele. He was at the MIT AI Lab concurrently with RMS. They even hacked together on EMACS, and he designed the original command set. He maintained the Jargon File before ESR did, and IMHO did a much better job. Instead of going on to found some movement to stroke his ego or impose his vision on everyone, however, he published important papers in computer science and has continued to work on standardizing and documenting programming languages, including C, Fortran, Common Lisp, Scheme, and Java. He's currently working on a replacement for Fortran called Fortress, which is pretty interesting even though it's not terribly relevant to me. Not all of it is earth-shaking stuff, but he's had (in my opinion, at least) a much stronger, albeit less visible, impact on computing than either ESR or RMS. It's guys like him that we should be looking up to and following in the footsteps of. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 07/16/2010 01:09 PM, Levi Pearson wrote: ESR annoys me almost as much as RMS. At least RMS was a real hacker; ESR is just kind of a hacker groupie with an inflated sense of self-importance. He's had some good ideas, he's made a few code contributions, and he's definitely been heavily involved in the periphery of the whole open source/free software movement, but my point is that the 'movement' aspect of software sharing and group development is useless and annoying. People naturally share ideas and help each other to work on projects. The early history of computer software 'hacker' culture shows this, and it worked pretty well without huge egos, figureheads, and organized movements. The fact that 'users groups' and the like have existed almost as long as computers have shows this, too. It worked just fine before RMS and ESR went on their crusades, and I believe we would have largely ended up in the same place we are now without the non-technical contributions of either of them. My experience has been different. More than a decade ago, Eric Raymond's papers convinced me to try something beyond Windows/Mac and take a serious look at Linux. I had to be convinced that free software and the people behind it were worth my time. Raymond's perspective made sense to me; Stallman's did not. So I think Raymond's writings were very valuable for inviting new people like me into the existing hacker culture. Consider yourself lucky that you were already part of that culture. Shane /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Rchard Stallman vs Darl McBride
On 07/14/2010 04:21 PM, Levi Pearson wrote: SNIP Seriously guys, this is a USERS group. Neither McBride nor Stallman would have anything useful to say about USING Linux. One of them is out to make a buck, and happened to end up on the wrong side of a lawsuit. The other is the leader of a social/political movement, and seems to be only really interested in advocating that movement. Have you had any personal interaction with Richard? I have and you are very wrong. He single-handedly created the entire movement that if it has not been done there would be no Linux for you to use. He also created the EMACS text editor. He also uses Linux. I joked to him that I would make sure that I would give him /usr/bin/emacs for his shell. You may disagree with his politics but without the FSF and the Gnu Public License (GPL) you would have no Linux OS to use. You would also not have GnuPG which is the free side of OpenPGP encryption that is on all Linux systems. From the looks of it some distros are using GnuPG instead of SSL certs now to make sure your downloads for the updates to your OS are legitimate. I would say Stallman is front and center behind the movement that gives you the Linux OS that you are using. The only importance Darl McBride has for me is as an example of how not to live your life by making an illegitimate and unethical money grab. In the past I was neutral about him. The more I learn about him the less I like him. He may be personable enough but he strikes me as both dumb and selfish. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On 7/15/2010 3:25 AM, Henry Hertz Hobbit wrote: On 07/14/2010 04:21 PM, Levi Pearson wrote: SNIP Seriously guys, this is a USERS group. Neither McBride nor Stallman would have anything useful to say about USING Linux. One of them is out to make a buck, and happened to end up on the wrong side of a lawsuit. The other is the leader of a social/political movement, and seems to be only really interested in advocating that movement. Have you had any personal interaction with Richard? I have and you are very wrong. He single-handedly created the entire movement that if it has not been done there would be no Linux for you to use. He also created the EMACS text editor. He also uses Linux. I joked to him that I would make sure that I would give him /usr/bin/emacs for his shell. You may disagree with his politics but without the FSF and the Gnu Public License (GPL) you would have no Linux OS to use. You would also not have GnuPG which is the free side of OpenPGP encryption that is on all Linux systems. From the looks of it some distros are using GnuPG instead of SSL certs now to make sure your downloads for the updates to your OS are legitimate. I would say Stallman is front and center behind the movement that gives you the Linux OS that you are using. The only importance Darl McBride has for me is as an example of how not to live your life by making an illegitimate and unethical money grab. In the past I was neutral about him. The more I learn about him the less I like him. He may be personable enough but he strikes me as both dumb and selfish. Richard Stallman would win with his GNU Lazer vision unless Darl McBride could use his patent infringement suit in time. Richard's getting on in years, so I think it would be close. No offense meant, unless you want to take some or something, but you're not exactly the forefront authority on social graces and human value in my mind. Your opinion actually makes me think less of Stallman and I'm more intrigued by McBride. -Tod Hansmann /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Thu, Jul 15, 2010 at 3:25 AM, Henry Hertz Hobbit hhhob...@securemecca.com wrote: On 07/14/2010 04:21 PM, Levi Pearson wrote: SNIP Seriously guys, this is a USERS group. Neither McBride nor Stallman would have anything useful to say about USING Linux. One of them is out to make a buck, and happened to end up on the wrong side of a lawsuit. The other is the leader of a social/political movement, and seems to be only really interested in advocating that movement. Have you had any personal interaction with Richard? I have and you are very wrong. He single-handedly created the entire movement that if it has not been done there would be no Linux for you to use. He also created the EMACS text editor. He also uses Linux. I joked to him that I would make sure that I would give him /usr/bin/emacs for his shell. You may disagree with his politics but without the FSF and the Gnu Public License (GPL) you would have no Linux OS to use. You would also not have GnuPG which is the free side of OpenPGP encryption that is on all Linux systems. From the looks of it some distros are using GnuPG instead of SSL certs now to make sure your downloads for the updates to your OS are legitimate. I would say Stallman is front and center behind the movement that gives you the Linux OS that you are using. I know very well who he is and how cool of a hacker he used to be. He created lots of cool stuff, and I'm really glad he did. But sometimes guys who created cool stuff decide they have another calling in life and move on. That's what RMS did. I'm not really interested in what he moved on to, and I don't think anything he's likely to say to a group is going to touch on the technical aspects of using Linux today. I'm sure he still uses it, but that's because it's what he knows and it's what fits his philosophy and social agenda. Linux would exist just fine without GNU, and if it didn't, we'd all be using BSD-derivatives instead. I was using Linux when it first took off, and the world was primed for something like it to take off. The only reason BSD didn't take the popularity surge that Linux did was a small and short-lived licensing issue that existed right at the cusp of things. I would be fine if BSD had taken off instead of Linux + GNU, really. We'd be largely in the same place, except maybe we'd have fewer Free Software zealots and religious wars over licenses. I'd love to hear him talk about his MIT AI Lab days and how the Lisp Machines there worked. I'd love to hear him talk about the development of emacs and how he uses it today. I'd love to hear about how he was involved in developing gcc and the technical details about how it's put together (I've actually had to fix a bad port of gcc to an embedded platform I was using, so something like that would have been very useful to me at one point). I'd love to hear him talk about any of the cool things he might have done if he hadn't dedicated his life to lecturing and advocating about Free Software. I really doubt he'd talk to a group about any of that stuff. He'd talk about Free Software and why he believes it's so important. That's fine that he's passionate about that now, but I don't really care and I don't think it's particularly relevant to helping people get stuff done with Linux or their computers in general. --Levi /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
Re: Rchard Stallman vs Darl McBride
On Thu, Jul 15, 2010 at 10:04 AM, Levi Pearson levipear...@gmail.com wrote: I know very well who he is and how cool of a hacker he used to be. He created lots of cool stuff, and I'm really glad he did. But sometimes guys who created cool stuff decide they have another calling in life and move on. That's what RMS did. I'm not really interested in what he moved on to, and I don't think anything he's likely to say to a group is going to touch on the technical aspects of using Linux today. I'm sure he still uses it, but that's because it's what he knows and it's what fits his philosophy and social agenda. Linux would exist just fine without GNU, and if it didn't, we'd all be using BSD-derivatives instead. I was using Linux when it first took off, and the world was primed for something like it to take off. The only reason BSD didn't take the popularity surge that Linux did was a small and short-lived licensing issue that existed right at the cusp of things. I would be fine if BSD had taken off instead of Linux + GNU, really. We'd be largely in the same place, except maybe we'd have fewer Free Software zealots and religious wars over licenses. I'd love to hear him talk about his MIT AI Lab days and how the Lisp Machines there worked. I'd love to hear him talk about the development of emacs and how he uses it today. I'd love to hear about how he was involved in developing gcc and the technical details about how it's put together (I've actually had to fix a bad port of gcc to an embedded platform I was using, so something like that would have been very useful to me at one point). I'd love to hear him talk about any of the cool things he might have done if he hadn't dedicated his life to lecturing and advocating about Free Software. I really doubt he'd talk to a group about any of that stuff. He'd talk about Free Software and why he believes it's so important. That's fine that he's passionate about that now, but I don't really care and I don't think it's particularly relevant to helping people get stuff done with Linux or their computers in general. --Levi I am interested in your ideas and wish to subscribe to your newsletter. /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */