* "proprietary", was Re: ada on linux
@ 2000-05-27 0:00 tmoran
2000-05-28 0:00 ` Robert Dewar
0 siblings, 1 reply; 39+ messages in thread
From: tmoran @ 2000-05-27 0:00 UTC (permalink / raw)
> > > Well most often proprietary in software has been used to
> > > describe software which is only released under restrictive
> > > license terms.
>
> > Such as requiring that changed source be released with changed
> > binaries ?
>
> No, such as not permitting redistribution at all, and not
> providing the sources at all.
"Anyone may use in any way" is not restrictive. "You must send money
to ___ if you give out copies to third parties" and "You must send source
code to ___ if you give out copies to third parties" are both restrictions.
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-27 0:00 "proprietary", was Re: ada on linux tmoran @ 2000-05-28 0:00 ` Robert Dewar 2000-05-28 0:00 ` Ken Garlington 2000-05-28 0:00 ` tmoran 0 siblings, 2 replies; 39+ messages in thread From: Robert Dewar @ 2000-05-28 0:00 UTC (permalink / raw) In article <u5XX4.297$q86.81131@news.pacbell.net>, tmoran@bix.com wrote: > "Anyone may use in any way" is not restrictive. "You must > send money > to ___ if you give out copies to third parties" and "You must send source > code to ___ if you give out copies to third parties" are both restrictions. I will repeat what I said earlier, I think there is a significant and important distinction between the position that you cannot redistribute under any circumstances, and the position that you can freely redistribute what you received! The above second restriction is *NOT* an accurate characterization of the GPL restrictions. If you get a copy of something under the GPL, you can freely redistribute it. To fulfill the requirement for distributing sources, you can simply point to the original distribution. For example, if you download the public binary version of GNAT, you can freely redistribute it to others, you do NOT have to distribute the sources yourself (the GPL never requires distribution of sources along with the objects, it just requires they be made available). So when you pass on the binaries, people can get the sources from the original location (e.g. cs.nyu.edu) and that's just fine. Tom, you seem to bat almost 100% when it comes to misinterpreting the GPL. Time to reread it carefully :-) Like any software license agreement, it needs careful reading! To be fair, you did not mention the GPL in the above, but I am willing to bet it is what you had in mind when you wrote: "You must send source code to ___ if you give out copies to third parties" The restriction that Tom alludes to above is a different one. It affects only derived works. Here the restriction is that if you create derived works (not permitted at all with typical proprietary software), then you can still distribute them, but you must distribute your mods in source form (e.g. as a patch file). Once again, with regard to the use of the word proprietary. There is a very big difference between Free Software that permits unrestricted redistribution and guarantees that the sources are always available, and typical software (e.g. from Microsoft or Ada vendors other than ACT) that significantly restricts or forbids redistribution. Usually people use the term proprietary software to refer to the latter. I don't really care what term people use, just so long as they recognize the distinction! People are still often confused by this distinction. Some months ago, I had to spend quite a bit of time on the phone explaining to a procurement guy at Lockheed that he did NOT need us to execute a source escrow agreement, because we had already supplied all the sources, and Lockheed had all the access that they would get from opening a source escrow (and more) already. Finally he understood, and agreed that "OK, I guess we don't need a source escrow in that case" :-) Robert Dewar Ada Core Technologiesh Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` Robert Dewar @ 2000-05-28 0:00 ` Ken Garlington 2000-05-28 0:00 ` Robert Dewar 2000-05-28 0:00 ` tmoran 1 sibling, 1 reply; 39+ messages in thread From: Ken Garlington @ 2000-05-28 0:00 UTC (permalink / raw) "Robert Dewar" <dewar@gnat.com> wrote in message news:8gppqa$og7$1@nnrp1.deja.com... > Some months > ago, I had to spend quite a bit of time on the phone explaining > to a procurement guy at Lockheed Lockheed Martin, please. The distinction is subtle but important. > that he did NOT need us to > execute a source escrow agreement, because we had already > supplied all the sources, and Lockheed had all the access that > they would get from opening a source escrow (and more) already. > Finally he understood, and agreed that "OK, I guess we don't > need a source escrow in that case" :-) A procurement guy who finally understood? OK, maybe it's not Lockheed Martin after all :) ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` Ken Garlington @ 2000-05-28 0:00 ` Robert Dewar 2000-05-30 0:00 ` Ted Dennison 0 siblings, 1 reply; 39+ messages in thread From: Robert Dewar @ 2000-05-28 0:00 UTC (permalink / raw) In article <gj_X4.2373$s64.80377@news.flash.net>, "Ken Garlington" <Ken.Garlington@computer.org> wrote: > Lockheed Martin, please. The distinction is subtle but > important. Indeed! Sorry for the excessive abbreviation in my post Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` Robert Dewar @ 2000-05-30 0:00 ` Ted Dennison 2000-05-30 0:00 ` Ken Garlington 0 siblings, 1 reply; 39+ messages in thread From: Ted Dennison @ 2000-05-30 0:00 UTC (permalink / raw) In article <8gq6r9$un$1@nnrp1.deja.com>, Robert Dewar <robert_dewar@my-deja.com> wrote: > In article <gj_X4.2373$s64.80377@news.flash.net>, > "Ken Garlington" <Ken.Garlington@computer.org> wrote: > > Lockheed Martin, please. The distinction is subtle but > > important. > > Indeed! Sorry for the excessive abbreviation in my post For some reason, those who came up from the Martin Marietta side of thengs tend to get touchy about that. Since I came from GE Aerospace portion of Martin, I never really cared. Silly as it sounds, inside the company they still keep track of who came from where. So to an insider "Lockheed" is short for "Lockheed heritage", and means "an LMC organization that used to be part of Lockheed". But to the rest of the world, its about as silly as GNU/Linux. :-) As for getting a bean-counter from LMC to understand anything that isn't along their well-trod path, I sympathise completely. Trying to introduce thought into a process that runs on the orderly flow of forms from one desk to another is an uphill battle. Plese don't confuse anyone here with a LMC paper-pusher though. Perhaps Tom hasn't fully grokked the GPL, but that doesn't put him in their league! -- T.E.D. http://www.telepath.com/~dennison/Ted/TED.html Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` Ted Dennison @ 2000-05-30 0:00 ` Ken Garlington 0 siblings, 0 replies; 39+ messages in thread From: Ken Garlington @ 2000-05-30 0:00 UTC (permalink / raw) "Ted Dennison" <dennison@telepath.com> wrote in message news:8h0hmg$95u$1@nnrp1.deja.com... > In article <8gq6r9$un$1@nnrp1.deja.com>, > Robert Dewar <robert_dewar@my-deja.com> wrote: > > In article <gj_X4.2373$s64.80377@news.flash.net>, > > "Ken Garlington" <Ken.Garlington@computer.org> wrote: > > > Lockheed Martin, please. The distinction is subtle but > > > important. > > > > Indeed! Sorry for the excessive abbreviation in my post > > For some reason, those who came up from the Martin Marietta side of > thengs tend to get touchy about that. Actually, I'm from the Lockheed (nee General Dynamics) side. I just know that the corporate position is to protect the trademark by not allowing the name to be shortened. Aviation Week ran a (somewhat snide) article some time back about the postcards that LM legal sends to publications that shorten the name... ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` Robert Dewar 2000-05-28 0:00 ` Ken Garlington @ 2000-05-28 0:00 ` tmoran 2000-05-28 0:00 ` David Starner 2000-05-28 0:00 ` Robert Dewar 1 sibling, 2 replies; 39+ messages in thread From: tmoran @ 2000-05-28 0:00 UTC (permalink / raw) >I will repeat what I said earlier, I think there is a >significant and important distinction between the position >that you cannot redistribute under any circumstances, and >the position that you can freely redistribute what you >received! I fully agree. How frequently does either position actually occur? I don't know of anybody who sells software with the requirement that you cannot redistribute under any circumstances. Even Microsoft allows retailers and system vendors to redistribute their software. The cirumstances, of course, include paying Microsoft royalties. It's true that I understand the GPL to say that you cannot "freely" redistribute what you received. As I (mis?) understand it, if you redistribute, there are certain requirements, ie, you are not free to do whatever you want. Perhaps the restriction that you must include a sentence pointing to where the source can be found is not an onerous one in the usual case, but it *is* a restriction on "freely". To construct a thought experiment, suppose you write a study guide for the SAT. To let your readers test themselves, you include on a CDROM the binary of a GPL'ed SAT simulation program. As I understand it, you haven't fulfilled the requirements. But if you include the source code, or even a pointer to it, some people are going to look at the source code, find out the questions and their correct answers, and then "ace" the simulated test. Now it may be that someone who does such a thing with a practice SAT test is a fool. Suppose we change the scenario slightly, and have the test be one for an on-line college course. Now the cheater will get credit for something he does not in fact know, and perhaps be hired in some position where an ignorant person can do significant damage. In the absence of the "source availability" requirement, you could have avoided this. If you used a non-GPL program, its author almost surely would be happy to let you redistribute, though he might want a small royalty per CDROM. If you offer a product that is, or uses, GPLed software, are you effectively required to charge for support? If you don't, and some users have made "just a little improvement", you face a heck of a lot of support time tracking down their "just a little mistake"s. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` tmoran @ 2000-05-28 0:00 ` David Starner 2000-05-28 0:00 ` Robert Dewar 1 sibling, 0 replies; 39+ messages in thread From: David Starner @ 2000-05-28 0:00 UTC (permalink / raw) On Sun, 28 May 2000 05:41:30 GMT, tmoran@bix.com <tmoran@bix.com> wrote: > To construct a thought experiment, suppose you write a study guide for >the SAT. To let your readers test themselves, you include on a CDROM the >binary of a GPL'ed SAT simulation program. As I understand it, you >haven't fulfilled the requirements. But if you include the source code, >or even a pointer to it, some people are going to look at the source code, >find out the questions and their correct answers, and then "ace" the Why would questions and answers be stored in the source code? >simulated test. Now it may be that someone who does such a thing with a >practice SAT test is a fool. Suppose we change the scenario slightly, and >have the test be one for an on-line college course. Now the cheater will >get credit for something he does not in fact know, and perhaps be hired in >some position where an ignorant person can do significant damage. In the >absence of the "source availability" requirement, you could have avoided this. No, you couldn't have. If you wrote the obvious code, the test will be extractable by running 'strings' over the binary. Even without that, you can still disassemble the code / watch the memory of the running program and get the test. If you give them the answers in any form, someone with enough patience will be able to get them, and if it actually matters someone will do it. Don't distribute the test. Run it over the net, then you don't have to send them the answers before hand. -- David Starner - dstarner98@aasaa.ofe.org http/ftp: x8b4e53cd.dhcp.okstate.edu Roleplaying Anonymous - a 1d12 step program ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` tmoran 2000-05-28 0:00 ` David Starner @ 2000-05-28 0:00 ` Robert Dewar 2000-05-28 0:00 ` tmoran 1 sibling, 1 reply; 39+ messages in thread From: Robert Dewar @ 2000-05-28 0:00 UTC (permalink / raw) In article <ek2Y4.340$q86.98777@news.pacbell.net>, tmoran@bix.com wrote: > I fully agree. How frequently does either position actually > occur? I don't know of anybody who sells software with the > requirement that you cannot redistribute under any > circumstances. Even Microsoft allows retailers and system > vendors to redistribute their software. The > cirumstances, of course, include paying Microsoft royalties. Seems like sophistry to me. The license does NOT permit any redistribution. Microsoft retailers do NOT redistribute copies in the sense we are talking about. Yes, system vendors can redistribute software under special licenses, but I was talking specifically about the typical license. FOr example, if you buy a copy of PowerPoint, you MAY NOT redistribute copies of this copy under the license you received, so yes, it happens all the time. In fact I assume that CLAW, being a proprietary product in this sense, also comes with a license that forbids unrestricted redistribution. > It's true that I understand the GPL to say that you cannot > "freely" redistribute what you receive Well you misunderstand, if you receive a software package under the GPL, you are absolutely free to redistribute what you receive. Restating your misunderstanding repeatedly does not change that it is a misunderstanding. After a while I have to wonder why you are insisting on repeating this false statement. The whole idea of the GPL in connection with freely distributed products like GNAT is not only to permit, but to encourage such secondary distribution. d. As I (mis?) understand it, if you > redistribute, there are certain requirements, ie, you are not > free to do whatever you want. In terms of redistributing the original that you received, assuming the original distribution was legal (i.e. conformed to the GPL), then yes, you can freely redistribute what you received. How many more times does this need to be said? It happens all the time with GNAT. People grab the file from cs.nyu.edu, and then send copies of that file to their friends or put it up on their own site. That is perfectly fine and perfectly in accordance with the GPL. As I said in my note, the requirement for distributing sources comes into play when deriviative works are created, a whole different ball game. > Perhaps the restriction that you must include a sentence > pointing to where the source can be found is not an onerous > one in the usual case, but it *is* a restriction on "freely". You really *insist* on this misunderstanding I guess. There is no requirement for you to include a sentence of this nature if you are simply redistributing what you received assuming that this information was present in the original. (which it most certainly should be). > To construct a thought experiment, suppose you write a study > guide for the SAT. To let your readers test themselves, you > include on a CDROM the > binary of a GPL'ed SAT simulation program. As I understand > it, you > haven't fulfilled the requirements. Let's assume you are NOT the original author of the simulation program here (the original author is of course not restricted in any way by the GPL, if you do not understand that, you are really at sea -- a license *I* issue cannot restrict me! Assuming that, the distribution of the GPL'ed SAT simulation program is a copyright violation on its face. <<irrelevant stuff snipped>> (irrelevant since it simply seems to be an argument that the GPL is not the right license to use in this situation. So what? What's that have to do with it? Have you seen me argue that all software should be distributed under the GPL? The answer is no you have not, since I have never made such a statement.) > If you offer a product that is, or uses, GPLed software, are > you effectively required to charge for support? AARGH! This is so far off base that I don't know what to say. I guess I will just say no and leave it at that > If you don't, and some > users have made "just a little improvement", you face a heck > of a lot of support time tracking down their "just a little > mistake"s. Again, completely out of left field. Just because you offer a product which is GPL'ed does not create any support obligations at all. If you do offer support, then the terms and conditions of the support will deal with the issue of modifications. This is not something new with the GPL, mainframe software has often been issued with full sources, with the understanding that the customer may (or even must) make changes, and then the support agreement has to spell out the obligations on both sides. In the case of GNAT, our support is premised on use of the binary versions that we distribute. In the cases where customers make changes, we continue to support on a best efforts basis, or in some cases, special support agreements are arranged. This is all really quite simple, I can't understand why you are making such heavy weather of it, unless your goal is simply to encourage people to agree that it is complex. Certainly I understand that you (or rather the company you work for) have decided to use restrictive licenses, and that is definitely your privilege, and may indeed be the best choice for you, since I know nothing about your business I could not comment. But the GPL model is quite simple, and works very well for us, and is really not that difficult to understand. These days we even find that procurement agents at companies like Lockheed Martin understand it quite fine :-) :-) Robert Dewar Ada Core Technologies Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` Robert Dewar @ 2000-05-28 0:00 ` tmoran 2000-05-28 0:00 ` David Starner ` (2 more replies) 0 siblings, 3 replies; 39+ messages in thread From: tmoran @ 2000-05-28 0:00 UTC (permalink / raw) >The license does NOT permit any redistribution. Microsoft I think you mean "the kind of license typically given to an end user allows him to run, but not redistribute, the program". And other licenses, typically given to retailers, allow redistribution, and still others given to system vendors may allow a variety of different things. The point is, your statement sounds like a general one about licenses (or at least "proprietary software" licenses) but in fact it's a specific statement about just one kind of license, and in that way misleading. >Well you misunderstand, if you receive a software package under >the GPL, you are absolutely free to redistribute what you If you receive a cake, are you free to redistribute it only in its entirety, or in slices? Does the GPL allow you the freedom to redistribute portions of what you received, for instance, just the binary? Does it allow you to examine the software, decide it has a very low probability of bugs, and then to redistribute it backed with a warranty? Your "absolutely free" is in fact a freedom with very definite, and sometimes onerous, restrictions, and in that way misleading. > > If you offer a product that is, or uses, GPLed software, are > > you effectively required to charge for support? > Again, completely out of left field. Just because you offer > a product which is GPL'ed does not create any support > obligations at all. If you do offer support, then the > terms and conditions of the support will deal with the > issue of modifications. Is it really practical to offer GPLed software with source code and support like "We offer no-cost support for this product, as long as you have made no changes. If you have changed anything, it makes it harder for us to help you, so you must pay $xxx for support. If you say you've made no changes, and we spend time helping you, and it turns out you have in fact made changes, then you must $yyy for the time we've spent." I suspect that in practice, ie, "effectively", you will find that you *must* charge for support, or you'll lose a lot of time and money giving free support to some people who create problems and then lie about the source of those problems. >GPL is not the right license to use in this situation. If Jones writes a program, can he offer it under different licensing arrangement to different people, one of those arrangements being the GPL? >Certainly I understand that you (or rather the company you >work for) have decided to use restrictive licenses, and that Calling other people's licenses "restrictive" and your own "absolutely free", is not, IMHO, a fair statement of the case. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` tmoran @ 2000-05-28 0:00 ` David Starner 2000-05-29 0:00 ` Robert Dewar 2000-05-30 0:00 ` Ted Dennison 2 siblings, 0 replies; 39+ messages in thread From: David Starner @ 2000-05-28 0:00 UTC (permalink / raw) On Sun, 28 May 2000 22:48:38 GMT, tmoran@bix.com <tmoran@bix.com> wrote: >support, or you'll lose a lot of time and money giving free >support to some people who create problems and then lie about the >source of those problems. Is this not true for any type of support? I've read enough horror stories about tech support on stuff that doesn't come with source to find this limited to stuff that comes with source. -- David Starner - dstarner98@aasaa.ofe.org http/ftp: x8b4e53cd.dhcp.okstate.edu Roleplaying Anonymous - a 1d12 step program ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` tmoran 2000-05-28 0:00 ` David Starner @ 2000-05-29 0:00 ` Robert Dewar 2000-05-29 0:00 ` tmoran 2000-05-30 0:00 ` Ted Dennison 2 siblings, 1 reply; 39+ messages in thread From: Robert Dewar @ 2000-05-29 0:00 UTC (permalink / raw) In article <anhY4.1121$M72.360470@news.pacbell.net>, tmoran@bix.com wrote: > >The license does NOT permit any redistribution. Microsoft > I think you mean "the kind of license typically given to an end > user allows him to run, but not redistribute, the program". > And > other licenses, typically given to retailers, allow > redistribution That's just completely wrong. You are inventing, and you are inventing wrong. Don't guess and publish it as fact :-) A retailer buys a copy of a program like Power Point. Now under 117, the retailer owns a copy of the progrma (not the program itself but a copy). The retailer is free under the copyright law to sell this copy to someone else (if you buy it, you inherit this right, and can then sell your unopened copy to someone else.) This right comes from the copyright law, not from any license. In fact the vendor never executes any license agreement with regard to this particular copy of Power Point (there may be a separate sales contract, but that's different). If you sell your unopened copy, you also never execute a license agreement (the shrink wrap license comes into effect only when the shring wrap is opened). > and still others given to system vendors may > allow a variety of different things. Yes, that's right, see the Microsoft trial for many details about such licenses. > The point is, your > statement > sounds like a general one about licenses (or at least "proprietary > software" licenses) but in fact it's a specific statement about > just one kind of license, and in that way misleading. I am talking about end user licenses, and have always made that clear. yes, system vendors do have other licenses. Retailers do not, that's just a Moran invention :-) > If you receive a cake, are you free to redistribute it only in > its entirety, For sure in its entirety, which is in practice the critical issue, and is what distinguishes it from proprietary products like CLAW. I realize that you are trying to minimize the difference here Tom, but it is really a VERY important difference, and trying to obfuscate this fact does not change the fact that there is for example a very big difference between CLAW and GtkAda here. > Does it allow you to examine the software, decide it has > a very low probability of bugs, and then to redistribute it > backed with a warranty? Yes, of course. Why are you asking questions like this whose answer should be obvious if you read the license. These questions seem designed to obfuscate. > Your "absolutely free" is in fact a freedom > with very definite Noone said you are absolutely free to do anything you want. That's an invention on your part. > and sometimes onerous, restrictions, and in > that way misleading. Onerous in one case in particular, which is the case where a proprietary software vendor wants to redistribute a derived work in proprietary form. For example, I quite see that the CLAW folks might like to use GtkAda to make a proprietary version of CLAW for non-NT targets. Well they definitely cannot do this. > Is it really practical to offer GPLed software with source > code > and support like "We offer no-cost support for this product, > as > long as you have made no changes. If you have changed anything, > it makes it harder for us to help you, so you must pay $xxx for > support. Yes, it is practical, and basically that is the situation at ACT, we support GNAT as we distribute it, and will fix any problems that are found. If you want us to guarantee to fix problems in versions you have modified, we can do this but we definitely write specialized (more expensive) support contracts in this place. What on earth would make you think it is impractical? It is the only approach that makes sense. > If you say you've made no changes, and we spend time > helping you, and it turns out you have in fact made changes, > then > you must $yyy for the time we've spent." I suspect that in > practice, ie, "effectively", you will find that you *must* charge for > support, or you'll lose a lot of time and money giving free > support to some people who create problems and then lie about > the source of those problems. You know nothing about distributing GPL'ed software and supporting it, because you have never done it. I suggest people take the actual experience of those who DO distribute GPL'ed software over the suspicions of someone with no experience at all. I quite understand the kind of arguments people like Tom (CLAW) use to justify choosing to go the proprietary route, but these arguments are often misinformed, as in this case. This does not mean that it is wrong to choose the proprietary route, but certainly uninformed suspicions should not be the basis for the decision. > >Certainly I understand that you (or rather the company you > >work for) have decided to use restrictive licenses, and that > Calling other people's licenses "restrictive" and your own > "absolutely free", is not, IMHO, a fair statement of the case. I understand how those who have chosen a proprietary route would like to minimize the differences, since they have a commercial interest in doing so, but in trying to make this argument, you should not exaggerate. Only *you* have called the GPL absolutely free. All I said was that one characteristic of the GPL is that you are absolutely free to redistribute what you received. By this I mean redistributing the entire product, I am not talking about deriviative works. It is definitely the case that the GPL is like the Microsoft License in that, for example, both ACT and MS are selling copies of copyright license with a license that places restrictions on the user. For more details on this, see the transcript of the GEAC vs GRACE trial in Newark Federal District Court, where I testified in detail on this point (the attorney for the other side being confused like Tom :-) The difference is in the qualitative nature of the restrictions. The reason that lawyers find the GPL odd as a license agreement is that it reads as being more interested in ensuring what the user CAN do than what they CANNOT do. If you can't see that CLAW has a much more restrictive license than GtkAda (I am choosing vaguely competing products here), then you are deluding yourself, and you will not serve your customers well by this delusion. As I said earlier, companies choose licenses to best meet their interests. No one ever said that the GPL is the right license for everything. In fact as you should know well, Ada Core Technologies considers the GPL to be unsuitable for a considerable chunk of the technology (including the Ada runtime and ASIS for example). Customers choose products on many factors, including the license. Some license provisions are important some are not. Actually for most of our customers, the redistribution right is not particularly relevant, big companies are not in the business of doing extra work to distribute copies of software to their potential competitors. For them, what is important about the GPL is the accessibility of the sources, and the open source nature of the technology in terms of development. To repeat, I have absolutely no criticism of the decision to use a proprietary license, by which I mean a license that is considerably more restrictive than the GPL, and in fact more akin to the Microsoft license, for CLAW. Many, in fact most of our tool partners (including the CLAW folks) choose to license their software in this manner. There are definitely arguments that can be made that this helps increase revenue to the company involved. Whether these are correct arguments or not depends on the circumstances. Certainly the automatic assumption that a proprietary license increases revenues is suspect. However, it is really not possible to construct an argument that a more restrictive license is beneficial to your users (other than perhaps arguing that it decreases costs). In terms of user benefits, a more permissive license is a clear win in all terms. Robert Dewar Ada Core Technologies Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-29 0:00 ` Robert Dewar @ 2000-05-29 0:00 ` tmoran 2000-05-29 0:00 ` Robert Dewar 0 siblings, 1 reply; 39+ messages in thread From: tmoran @ 2000-05-29 0:00 UTC (permalink / raw) Thank you for clarifying things: > > a very low probability of bugs, and then to redistribute it > > backed with a warranty? > Yes, of course. Why are you asking questions like this whose > answer should be obvious if you read the license. Sorry, I had looked at the license included in the Gnat library sources, and not at the line in the NO WARRANTY section of the FSF GPL that does indeed say the parties may contract in writing for a warranty. >Noone said you are absolutely free to do anything you want. >I am talking about end user licenses, and have always made >that clear. > and support like "We offer no-cost support for this product, >... >ACT, we support GNAT as we distribute it, and will fix any >By this I mean redistributing the entire product, >both ACT and MS are selling copies of copyright license with a >license that places restrictions on the user. I'm glad you've now made those things clear. >I quite understand the kind of arguments people like Tom (CLAW) You misunderstand here. Tom is not CLAW or vice versa. Tom has worked on CLAW on occasion for RR Software, but is not an employee, not an executive, and most certainly not a person with authority over RR Software licensing terms. Tom has, however, read an economics book and recognizes how a monopolist may increase profits by setting different prices for different customers. >Certainly the automatic assumption that a proprietary license >increases revenues is suspect. Agreed. "Give away the razor, charge for the blades" often brings greater revenues. >In terms of user benefits, a more permissive license is a clear >win in all terms. Agreed. But a *differently permissive* license may or may not be a win, depending on the particular user. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-29 0:00 ` tmoran @ 2000-05-29 0:00 ` Robert Dewar 0 siblings, 0 replies; 39+ messages in thread From: Robert Dewar @ 2000-05-29 0:00 UTC (permalink / raw) In article <9qBY4.1017$vx2.469023@news.pacbell.net>, tmoran@bix.com wrote: > Sorry, I had looked at the license included in the Gnat > library sources, > and not at the line in the NO WARRANTY section of the FSF GPL > that does > indeed say the parties may contract in writing for a warranty. Be careful that the "license" in the GNAT library sources is NOT the full text of the license, it incorporates by reference the entire GPL, and then modifies this text. Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-28 0:00 ` tmoran 2000-05-28 0:00 ` David Starner 2000-05-29 0:00 ` Robert Dewar @ 2000-05-30 0:00 ` Ted Dennison 2000-05-30 0:00 ` Robert Dewar 2 siblings, 1 reply; 39+ messages in thread From: Ted Dennison @ 2000-05-30 0:00 UTC (permalink / raw) In article <anhY4.1121$M72.360470@news.pacbell.net>, tmoran@bix.com wrote: > If Jones writes a program, can he offer it under different > licensing arrangement to different people, one of those > arrangements being the GPL? If "Jones" is the sole holder of the copyright on said program, yes. To avoid this potential route of imprisoning software for the Linux kernel, Linus makes sure that everyone who makes a significant contribution to a kernel source gets a piece of the copyright. That way if someone wanted the Linux kernel distributed under a different license, they'd have to get the permission (iow: pay off) every copyright holder for every kernel source file. I took the same approach with OpenToken, for the same reason. > >Certainly I understand that you (or rather the company you > >work for) have decided to use restrictive licenses, and that > Calling other people's licenses "restrictive" and your own > "absolutely free", is not, IMHO, a fair statement of the case. In principle, perhaps not. In this particular case, yes it is (unless RR's license has changed quite a bit from the last time I looked at it). I highgly suggest you go up to www.gnu.org and look through the relevent documents. The definition of "Free Software" is at http://www.gnu.org/philosophy/free-sw.html . It would be hard to argue that RR's software fits this defintion better than ACT's. An excerpt: ----- ``Free software'' refers to the users' freedom to run, copy, distribute, study, change and improve the software. More precisely, it refers to four kinds of freedom, for the users of the software: The freedom to run the program, for any purpose (freedom 0). The freedom to study how the program works, and adapt it to your needs (freedom 1). Access to the source code is a precondition for this. The freedom to redistribute copies so you can help your neighbor (freedom 2). The freedom to improve the program, and release your improvements to the public, so that the whole community benefits. (freedom 3). Access to the source code is a precondition for this. --- Also, you really should read the GPL. Its much shorter and eaiser to follow than this thread. :-) Any good software engineer should be capable of fully understanding it if they read it as they would a new API; with an eye towards what it allows you to do and doesn't allow you to do. The URL is: http://www.gnu.org/copyleft/gpl.html -- T.E.D. http://www.telepath.com/~dennison/Ted/TED.html Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` Ted Dennison @ 2000-05-30 0:00 ` Robert Dewar 2000-05-30 0:00 ` bill 2000-05-30 0:00 ` Ted Dennison 0 siblings, 2 replies; 39+ messages in thread From: Robert Dewar @ 2000-05-30 0:00 UTC (permalink / raw) In article <8h0jgj$amb$1@nnrp1.deja.com>, Ted Dennison <dennison@telepath.com> wrote: > In article <anhY4.1121$M72.360470@news.pacbell.net>, > tmoran@bix.com wrote: > > > If Jones writes a program, can he offer it under different > > licensing arrangement to different people, one of those > > arrangements being the GPL? > > If "Jones" is the sole holder of the copyright on said > program, yes. That's entirely correct. It is one special case of the fact that if you as copyright holder issue a license to someone, it does not in anyway restrict what YOU can do. > To avoid this potential route of imprisoning software for the > Linux kernel, Linus makes sure that everyone who makes a > significant contribution to a > kernel source gets a piece of the copyright. Note that another interesting way of achieving this is by contract. In the case where a copyright assignment is made to the FSF, the instrument of assignment includes a requirement that the FSF make the software available under the GPL for ever, so that they do NOT have the rights the original copyright holder would have. This is relevant to CLA, since the copyright for the basic GNAT technology was assigned from New York University to FSF. > That way if someone wanted the Linux kernel distributed under > a different license, they'd have to get the permission (iow: > pay off) every copyright holder for every > kernel source file. Or argue that a particular copyright holder does not in fact have a right to protectable code. After all Altai contains quite a long list of exclusions from what is protectable. Yes I know that's only third circuit, but in practice it is widely recognized. Also there are drawbacks to the multiple copyright holders situation. Has each of these copyright holders specifically licensed his work under the GPL. Note that merely putting a notice in a source file may or may not constitute granting such a license. Consider the following situation. Party A holds the copyright to some code X and releases it under the GPL. Party B makes a useful modification. He holds the copyright on this modification. He posts on the net "I have a really nice modification here to the code X" That does NOT mean that anyone is free to use this modification unless party B very explicitly releases the modification under the GPL. What is the status of the modified software in B's posession? Simple, it is a deriviative work, certainly permitted under the GPL, in which there are two interests A's and B's. The rub is that unless B agrees to the GPL licensing, the deriviative work simply cannot be distributed. One can even imagine B hiring himself out to make the same change to individual copies of program A, retaining full rights to this change. So if you go the multiple copyright holder route, be absolutely sure that you have proper legal instruments signed by every contributor. The requirement here is no different than if a copyright assignment was required. > I took the same approach with OpenToken, for the same reason. Have you been careful to get proper legal documents signed by all contributors. If not, it may well be the case that you are distributing something that cannot in fact be legally redistributed. Legal stuff gets complicated fast :-) Robert Dewar Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` Robert Dewar @ 2000-05-30 0:00 ` bill 2000-05-31 0:00 ` Florian Weimer 2000-06-03 0:00 ` Robert Dewar 2000-05-30 0:00 ` Ted Dennison 1 sibling, 2 replies; 39+ messages in thread From: bill @ 2000-05-30 0:00 UTC (permalink / raw) Why deal with all this mess of GPL, when one can download freeBSD which is all free? Bill ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` bill @ 2000-05-31 0:00 ` Florian Weimer 2000-06-01 0:00 ` Geoff Bull 2000-06-03 0:00 ` Robert Dewar 1 sibling, 1 reply; 39+ messages in thread From: Florian Weimer @ 2000-05-31 0:00 UTC (permalink / raw) bill@new writes: > Why deal with all this mess of GPL, when one can download > freeBSD which is all free? Does FreeBSD have an Ada compiler which is "all free"? What about a C compiler? ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-31 0:00 ` Florian Weimer @ 2000-06-01 0:00 ` Geoff Bull 2000-06-03 0:00 ` Robert Dewar 0 siblings, 1 reply; 39+ messages in thread From: Geoff Bull @ 2000-06-01 0:00 UTC (permalink / raw) This has got be the most pointless thread ever on CLA! ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-01 0:00 ` Geoff Bull @ 2000-06-03 0:00 ` Robert Dewar 0 siblings, 0 replies; 39+ messages in thread From: Robert Dewar @ 2000-06-03 0:00 UTC (permalink / raw) In article <39362445.3C630862@research.canon.com.au>, Geoff Bull <geoff@research.canon.com.au> wrote: > > This has got be the most pointless thread ever on CLA! I don't agree, since GNAT is an important artifact on the Ada scene, understanding what its status and licensing are is of interest to quite a lot of people. I do agree that some of the posts have been less than helpful, but the thread contains useful information for many (I even got a nice private thankyou note from someone for one of my explanatory posts :-) Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` bill 2000-05-31 0:00 ` Florian Weimer @ 2000-06-03 0:00 ` Robert Dewar 2000-06-03 0:00 ` tmoran 1 sibling, 1 reply; 39+ messages in thread From: Robert Dewar @ 2000-06-03 0:00 UTC (permalink / raw) In article <8h1747$i4a@drn.newsguy.com>, bill@new wrote: > > Why deal with all this mess of GPL, when one can download > freeBSD which is all free? Assuming this is not meant as a troll, which might well be the case, then the answer is straightforward. Authors choose to use the GPL to ensure that derived works remain all free. The trouble with freeBSD, is there is nothing to stop R. Dewar or anyone else making nonfreeveryexpensiveproprietaryBSD with some small changes. We want to ensure that all future versions of GNAT remain free! There is no "mess" with GPL, though I agree some of the obfuscation in this thread could be misinterpreted in this manner. The GPL is really quite clear, and quite effective. Usually I only see complaints or worries about the GPL from those associated with manufacture of proprietary software. Those people would of course prefer a BSD like license to the GPL license, because then they can use the software in their proprietary products. Let's take a specific example. At one point Tartan suggested the possibility of joining the GNAT front end with their proprietary backend to make a proprietary result. We told them they couldn't and that was that! On the other hand, we know of important cases where major manufacturers are open sourcing important code generators so that they can properly use GPL'ed front ends. Anyway, the important thing to remember is that the author of software decides on the licensing, and the only reasonable response anyone else can have to a license is to either decide that you like the software and license, and acquire it, or that you don't and decline! Robert Dewar Ada Core Technologies Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-03 0:00 ` Robert Dewar @ 2000-06-03 0:00 ` tmoran 2000-06-03 0:00 ` Dale Stanbrough ` (2 more replies) 0 siblings, 3 replies; 39+ messages in thread From: tmoran @ 2000-06-03 0:00 UTC (permalink / raw) >Usually I only see complaints or worries about the GPL from >those associated with manufacture of proprietary software. One would hardly expect people using the GPL to be frequent complainants. >Let's take a specific example. At one point Tartan suggested >the possibility of joining the GNAT front end with their >proprietary backend to make a proprietary result. We told >them they couldn't and that was that! So there could have been an Ada compiler with the advantages of both the Gnat front-end, and Tartan's back-end, but you said no. Was that because you didn't have the rights to do it, or because you didn't want Tartan to offer such a compiler? >... those associated with manufacture of proprietary software. >Those people would of course prefer a BSD like license to >the GPL license, because then they can use the software in >their proprietary products. Do you know as fact that all people who object to the GPL would prefer the BSD like license? Do you know as fact that's because then they would be able to use the software in their proprietary products? Or are you just guessing that anyone who disagrees with the OSF hype does so solely because it prevents them from thievery? ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-03 0:00 ` tmoran @ 2000-06-03 0:00 ` Dale Stanbrough 2000-06-03 0:00 ` Jeff Creem 2000-06-05 0:00 ` "proprietary", was Re: ada on linux Robert Dewar 2 siblings, 0 replies; 39+ messages in thread From: Dale Stanbrough @ 2000-06-03 0:00 UTC (permalink / raw) tmoran@bix.com wrote: > So there could have been an Ada compiler with the advantages of both > the Gnat front-end, and Tartan's back-end, but you said no. The "saying no" would have been more along the lines of "that would violate the copyrights for this software which is owned by the FSF" than "we are not going to let you do it" Dale ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-03 0:00 ` tmoran 2000-06-03 0:00 ` Dale Stanbrough @ 2000-06-03 0:00 ` Jeff Creem 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` "proprietary", was Re: ada on linux Robert Dewar 2 siblings, 1 reply; 39+ messages in thread From: Jeff Creem @ 2000-06-03 0:00 UTC (permalink / raw) > >... those associated with manufacture of proprietary software. > >Those people would of course prefer a BSD like license to > >the GPL license, because then they can use the software in > >their proprietary products. > Do you know as fact that all people who object to the GPL would > prefer the BSD like license? Do you know as fact that's because then > they would be able to use the software in their proprietary products? > Or are you just guessing that anyone who disagrees with the OSF hype > does so solely because it prevents them from thievery? I know in cases I have been concerned about, I have had to stay away from libraries that are pure GPL (not talking about GNAT here) for my projects since there are projects where portions of the source code are classified (by the government). We certainly give all of the source the the government and they in turn are free to give it to whomever they determine has the clearence and need to know. But, the clause in the GPL that says all third parties must have all the licence rights has always meant to me (and I am not a lawyer and I could be very wong here) that if some third party approached the company and said that they wanted a copy of program X from us that we would have to give it to them (but we could charge a fee).. Perhaps this is just another misconception..A while back a sent e-mail to FSF asking specific questions about the GPL in this and other areas...I got some answers but the rest of the reply said these are hard questions and I got no answers... We do of course have our own legal department that could look at these (and has) but I am often concerned that they give me answers for what the believe they can get away with based on the licenses and not what the original intent of the license is. (again another personal belief that is probably wong). Jeff ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-03 0:00 ` Jeff Creem @ 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` Jeff Creem 0 siblings, 1 reply; 39+ messages in thread From: Robert Dewar @ 2000-06-05 0:00 UTC (permalink / raw) In article <8haraq$k3j$1@pyrite.mv.net>, "Jeff Creem" <jcreem@atetola.mv.com> wrote: > I know in cases I have been concerned about, I have had to stay away > from libraries that are pure GPL (not talking about GNAT here) for my > projects > since there are projects where portions of the source code are classified > (by > the government). Absolutely! An author who decides to use the unmodified GPL is most definitely making the decision that you CANNOT use the software in this manner without negotiating a separate license. And that's the author's right under current copyright law. > that if > some third party approached the company and said that they wanted a copy > of program X from us that we would have to give it to them (but we could charge a fee).. This is complete nonsense. But it is a VERY common misconception. The GPL never forces you to distribute a program. If ACT suddenly decided to make no more public versions available, that would be perfectly in accord with the GPL requirements. As I have said many times before the fact that ACT makes public versions available is not required by the GPL, it is simply something that ACT decides to do for the general good of the Ada community. > Perhaps this is just another misconception..A while back a > sent e-mail to FSF asking specific > questions about the GPL in this and other areas...I got some > answers but > the rest of the > reply said these are hard questions and I got no answers... There is absolutely NO reason to expect the FSF to provide you with free legal services. The fact that they did answer some of your questions is entirely up to them. > We do of course have our own legal department that could look at these (and > has) but > I am often concerned that they give me answers for what the believe they can > get > away with based on the licenses and not what the original intent of the > license is. (again > another personal belief that is probably wong). You always have to ask your own lawyers what you can and cannot do under a license if it is not clear to you! Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-05 0:00 ` Robert Dewar @ 2000-06-05 0:00 ` Jeff Creem 2000-06-06 0:00 ` GPL distribution rules (was: "proprietary") Larry Kilgallen 0 siblings, 1 reply; 39+ messages in thread From: Jeff Creem @ 2000-06-05 0:00 UTC (permalink / raw) -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 First let me start off by saying that I only jumped into this because I find the whole area of software licensing interesting (apparently I am getting old and bizzare or something). So in my prior postings (and Roberts response) we have : > In article <8haraq$k3j$1@pyrite.mv.net>, > "Jeff Creem" <jcreem@atetola.mv.com> wrote: > > > I know in cases I have been concerned about, I have had to > stay away > > from libraries that are pure GPL (not talking about GNAT here) > for my > > projects > > since there are projects where portions of the source code are > classified > > (by > > the government). > > Absolutely! An author who decides to use the unmodified GPL > is most definitely making the decision that you CANNOT use > the software in this manner without negotiating a separate > license. And that's the author's right under current > copyright law. Ok..Seems fair..And it is exactly what I always assumed one of the ramifications of the unmodified GPL was (and again I understand that an author is well within their rights to set these limits).. But then curiously to one of my misconception ramblings we have this: > > > that if > > some third party approached the company and said that they > wanted a copy > > of program X from us that we would have to give it to them > (but we could charge a fee).. > > This is complete nonsense. But it is a VERY common > misconception. The GPL never forces you to distribute > a program. If ACT suddenly decided to make no more public > versions available, that would be perfectly in accord > with the GPL requirements. As I have said many times before > the fact that ACT makes public versions available is not > required by the GPL, it is simply something that ACT decides > to do for the general good of the Ada community. Hmm..But if that were the case then my first problem would be no problem at all since I would have given source and binaries to a customer who owns the secrets and I dont have to give it to anyone else and neither do they.. My common misconception problably comes from the following text from the GPL " 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.) " So..If I ever once send someone a binary without source code sure seems to me that for the next three years I must give source to any third party for a nominal fee... > > > Perhaps this is just another misconception..A while back a > > sent e-mail to FSF asking specific > > questions about the GPL in this and other areas...I got some > > answers but > > the rest of the > > reply said these are hard questions and I got no answers... > > There is absolutely NO reason to expect the FSF to provide > you with free legal services. The fact that they did answer > some of your questions is entirely up to them. I agree..And my comment was not meant to say disparage the FSF in any way however the FSF wants to promote "free" software. There are theses licenses GPL and LGPL and based on discussions I have seen I would bet that more than half of the people who put source out there under the GPL or LGPL have no idea what terms they are actually releasing the source under...Granted this is their problem.. Note that when I sent e-mail to them asking for help I did offer to send them a contribution (of course it would not have been enough to pay for a lawyer since I was looking into this for personal reasons). > > > We do of course have our own legal department that could look > at these (and > > has) but > > I am often concerned that they give me answers for what the > believe they can > > get > > away with based on the licenses and not what the original > intent of the > > license is. (again > > another personal belief that is probably wong). > > You always have to ask your own lawyers what you can and cannot > do under a license if it is not clear to you! > Again my point here is that lawyers will tell me what the license says I can do which is important from a legal point of view (and for example, I neither could nor would use something if my legal department said I could not)...But there is another issue here which is intent...It may be that the specific wording of a license allows something but it could be that the author never intended for it to allow it...For lawsuits I am interested in the former.. For sleeping at night and looking in the mirror I am interested in the latter. Jeff -----BEGIN PGP SIGNATURE----- Version: PGPfreeware 6.5.3 for non-commercial use <http://www.pgp.com> iQA/AwUBOTxJiS7VRsN7h5LhEQKJPACgzjDM+RVlkdeFe3h2HPHSwuDhiQoAoJTn 4anPf5Nkg+eVkdhngS8ji/1V =uSnz -----END PGP SIGNATURE----- ^ permalink raw reply [flat|nested] 39+ messages in thread
* GPL distribution rules (was: "proprietary") 2000-06-05 0:00 ` Jeff Creem @ 2000-06-06 0:00 ` Larry Kilgallen 0 siblings, 0 replies; 39+ messages in thread From: Larry Kilgallen @ 2000-06-06 0:00 UTC (permalink / raw) In article <8hhhej$mp2$1@pyrite.mv.net>, "Jeff Creem" <jcreem@atetola.mv.com> writes: > 3. You may copy and distribute the Program (or a work based on it, > under Section 2) in object code or executable form under the terms of > Sections 1 and 2 above provided that you also do one of the > following: > > a) Accompany it with the complete corresponding machine-readable > source code, which must be distributed under the terms of Sections 1 > and 2 above on a medium customarily used for software interchange; I don't see any requirement around compilers. Doesn't that leave the "loophole" (from the FSF viewpoint) that someone could modify some (non-compiler) software obtained under this license to use special language features not generally supported in the source language and release the modified source but not a suitable compiler ? ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-03 0:00 ` tmoran 2000-06-03 0:00 ` Dale Stanbrough 2000-06-03 0:00 ` Jeff Creem @ 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` Geoff Bull 2000-06-05 0:00 ` tmoran 2 siblings, 2 replies; 39+ messages in thread From: Robert Dewar @ 2000-06-05 0:00 UTC (permalink / raw) In article <sv1_4.104$%a3.95362@news.pacbell.net>, tmoran@bix.com wrote: > Do you know as fact that all people who object to the GPL > would prefer the BSD like license? Do you know as fact that's > because then they would be able to use the software in their > proprietary products? a) Yes, this corresponds to my experience in these discussions. b) Yes, it makes sense, since that's really the only major issue in practice. > Or are you just guessing that anyone who disagrees with the > OSF hype OSF? [what does OSF have to do this with this discussion??) hype? (what hype are you talking about here??) what on earth are you talking about here. This discussion is just about what various licenses permit. > does so solely because it prevents them from thievery? Who said anything about theivery? That's very very odd rhetoric. We are just talking about licenses here and what they permit. I really have no idea what you are talking about here either! Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-05 0:00 ` "proprietary", was Re: ada on linux Robert Dewar @ 2000-06-05 0:00 ` Geoff Bull 2000-06-05 0:00 ` tmoran 1 sibling, 0 replies; 39+ messages in thread From: Geoff Bull @ 2000-06-05 0:00 UTC (permalink / raw) Robert Dewar wrote: > > OSF? FSF? ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-05 0:00 ` "proprietary", was Re: ada on linux Robert Dewar 2000-06-05 0:00 ` Geoff Bull @ 2000-06-05 0:00 ` tmoran 2000-06-05 0:00 ` Geoff Bull 2000-06-05 0:00 ` Robert Dewar 1 sibling, 2 replies; 39+ messages in thread From: tmoran @ 2000-06-05 0:00 UTC (permalink / raw) >Those people would of course prefer a BSD like license to >the GPL license, because then they can use the software in >their proprietary products. If the software is GPLed, it can be used in "Those people"'s proprietary product simply by agreement of the copyright holder, since he can release his software under different licenses to different people. So the only difference is that with BSD "Those people" do not need permission of the copyright holder. While legal, adding significant value to one's proprietary product by copying/including something BSDed, without its author's permission, is usually considered unethical. So saying that all people who object to the GPL would prefer the BSD like license because then they would be able to use the software, without it's author's permission, in their proprietary products, is saying that all people who object to the GPL do so because they wish to engage in that kind of unethical behavior. That is false. There are many people who are not in fact interested in stealing other people's software, but still believe the GPL and its associated rhetoric very often have very negative consequences for both developers and users. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-05 0:00 ` tmoran @ 2000-06-05 0:00 ` Geoff Bull 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` Robert Dewar 1 sibling, 1 reply; 39+ messages in thread From: Geoff Bull @ 2000-06-05 0:00 UTC (permalink / raw) tmoran@bix.com wrote: > While legal, adding significant value to one's > proprietary product by copying/including something BSDed, without its > author's permission, is usually considered unethical. On what do you base this assertion ??? There is absolutely nothing to suggest this in the BSD license. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-05 0:00 ` Geoff Bull @ 2000-06-05 0:00 ` Robert Dewar 0 siblings, 0 replies; 39+ messages in thread From: Robert Dewar @ 2000-06-05 0:00 UTC (permalink / raw) In article <393B554F.43C369C4@research.canon.com.au>, Geoff Bull <geoff@research.canon.com.au> wrote: > tmoran@bix.com wrote: > > While legal, adding significant value to one's > > proprietary product by copying/including something BSDed, without its > > author's permission, is usually considered unethical. > > On what do you base this assertion ??? > There is absolutely nothing to suggest this in the BSD > license. Exactly, and indeed one of the main reasons that some authors prefer to use the BSD is precisely that they want everyone to be free to use the software without any restrictions. That's perfectly legitimate, and may indeed make a lot of sense in a particular situation. For example, when you buy the Palm Developer's Kit, you get a completely unfettered license to the source of the Palm utilities. Palm *wants* you to make improved software to sell more Palms! The choice of license really depends on the situation. For the GNAT runtime, we think it is a good thing if developers can freely use this code, and that is why we follow the pattern established by the FSF in similar situations to ensure that runtime code can be freely used. For the GNAT compiler, we think it is a good thing for the Ada community if we all have a guarantee that all future versions of GNAT will remain in open source/free software status, and that's why we choose to use the GPL for this part of the technology. Robert Dewar Ada Core Technologies Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-05 0:00 ` tmoran 2000-06-05 0:00 ` Geoff Bull @ 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` tmoran 1 sibling, 1 reply; 39+ messages in thread From: Robert Dewar @ 2000-06-05 0:00 UTC (permalink / raw) In article <CQH_4.1477$C02.345830@news.pacbell.net>, tmoran@bix.com wrote: > While legal, adding significant value to one's > proprietary product by copying/including something BSDed, > without its > author's permission, is usually considered unethical. This is complete nonsense. If the author releases something under the BSD, they are giving you *EXPLICIT* permission to "add significant value ...", and there is absolutely NOTHING unethical, inappropriate, or even unexpected about such action. The whole idea of the BSD license is to give such permission, and there are definitely authors who feel this is appropriate and is what they want, and that is why they use the license. If an author wants to be asked for permission for such actions, then they should not use the BSD license, it is as simple as that. > So saying that all > people who object to the GPL would prefer the BSD like license because > then they would be able to use the software, without it's author's > permission, in their proprietary products, is saying that all people who > object to the GPL do so because they wish to engage in that kind of > unethical behavior. Except that it is not unethical except in your confused thinking. > That is false. There are many people who are not in > fact interested in stealing other people's software, To think that using BSD software in this way is stealing is completely absurd, and actually really rather inappropriate. You are accusing people of a crime, where there is no crime, just misunderstanding on your part. Really Tom, if you are going to make such pronouncements, you must make more effort to understand how copyright and licensing works! > but still believe the > GPL and its associated rhetoric very often have very negative > consequences for both developers and users. What "very negative consequences"? Developers use this license if that's how they want to license their software, users acquire the software if they find the license conditions acceptable. This is the same situation as for any other licensed software. The only negative consequence is that a particular user or developer finds that they cannot make use of the software under some situation because the license is unsuitable. Fine, go find another product, that's what competition is about! Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-06-05 0:00 ` Robert Dewar @ 2000-06-05 0:00 ` tmoran 0 siblings, 0 replies; 39+ messages in thread From: tmoran @ 2000-06-05 0:00 UTC (permalink / raw) >The trouble with freeBSD, is there is nothing to stop R. Dewar > or anyone else making nonfreeveryexpensiveproprietaryBSD with > some small changes. That you find this perfectly acceptable is not the issue - your belief that this is the only reason people object to the GPL, is incorrect. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` Robert Dewar 2000-05-30 0:00 ` bill @ 2000-05-30 0:00 ` Ted Dennison 2000-05-30 0:00 ` Robert Dewar 1 sibling, 1 reply; 39+ messages in thread From: Ted Dennison @ 2000-05-30 0:00 UTC (permalink / raw) In article <8h0mt3$d5j$1@nnrp1.deja.com>, Robert Dewar <robert_dewar@my-deja.com> wrote: > Consider the following situation. Party A holds the copyright > to some code X and releases it under the GPL. > > Party B makes a useful modification. He holds the copyright > on this modification. He posts on the net > > "I have a really nice modification here to the code X" > > That does NOT mean that anyone is free to use this modification > unless party B very explicitly releases the modification under > the GPL. Well, I don't go and grab random code off the net to put in the baseline. Everything in OpenToken was either written by me or was submitted to me via (acrhived) email for the express purpose of being incorporated into the GMGPL baseline. But it certainly couldn't hurt to formailze this process a bit more. I'd hate for there to be a misunderstanding. -- T.E.D. http://www.telepath.com/~dennison/Ted/TED.html Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` Ted Dennison @ 2000-05-30 0:00 ` Robert Dewar 2000-05-30 0:00 ` About AdaOS Didier Utheza 2000-05-30 0:00 ` "proprietary", was Re: ada on linux Ted Dennison 0 siblings, 2 replies; 39+ messages in thread From: Robert Dewar @ 2000-05-30 0:00 UTC (permalink / raw) In article <8h0qd0$g1q$1@nnrp1.deja.com>, > Well, I don't go and grab random code off the net to put in the > baseline. Everything in OpenToken was either written by me or was > submitted to me via (acrhived) email for the express purpose of being > incorporated into the GMGPL baseline. But it certainly couldn't hurt to > formailze this process a bit more. I'd hate for there to be a > misunderstanding. Incorporating something into a GPL'ed program does not make the incorporation GPL'ed. That is a common misconception, but it is just that a misconception. The issue is one of redistribution. A derived work that contains GPL'ed code can (under the rights obtained from the GPL) only be distributed if all the other code has no more restrictive licensing requirements. So unless you have a very definite statement, preferably in writing signed, an email message is unlikely to be taken as a formal contract, or at least authenticity questions can easily be raised in this context, you may be in trouble. The statement must CLEARLY give the author's permission for incorporating the code under whatever licence you are using. There is a lot of sloppiness in these kind of procedures with code wandering around, but that sloppiness can come back to haunt you in court later on. Suppose someone offers you ten million dollars for a copy of OPENTOKEN with all these changes. You might well find that some of the other contributors object to getting $0. Yes, that is far fetched, and of course in the case of OPENTOKEN, there is unlikely to be a scenario in which the legal issues are ever explored, but that does not mean they do not exist. And Linux is after all quite another kettle of fish when it comes to $$$$, I can easily imagine some of the copyright holders for Linux getting persnickety if things have not been done exactly right :-) Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
* About AdaOS 2000-05-30 0:00 ` Robert Dewar @ 2000-05-30 0:00 ` Didier Utheza [not found] ` <WCBZ4.4122$XX4.63232@news-east.usenetserver.com> 2000-05-30 0:00 ` "proprietary", was Re: ada on linux Ted Dennison 1 sibling, 1 reply; 39+ messages in thread From: Didier Utheza @ 2000-05-30 0:00 UTC (permalink / raw) Is the AdaOS site stillonline. I try http://www.AdaOS.org/ and I get a forbidden access message. Is the access restricted or the site offline? Didier Utheza ^ permalink raw reply [flat|nested] 39+ messages in thread
[parent not found: <WCBZ4.4122$XX4.63232@news-east.usenetserver.com>]
* Re: About AdaOS [not found] ` <WCBZ4.4122$XX4.63232@news-east.usenetserver.com> @ 2000-06-01 0:00 ` Didier Utheza 0 siblings, 0 replies; 39+ messages in thread From: Didier Utheza @ 2000-06-01 0:00 UTC (permalink / raw) To: David Botton Thanks, I did. Very happy that the project is still on. Didier Utheza. ^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: "proprietary", was Re: ada on linux 2000-05-30 0:00 ` Robert Dewar 2000-05-30 0:00 ` About AdaOS Didier Utheza @ 2000-05-30 0:00 ` Ted Dennison 1 sibling, 0 replies; 39+ messages in thread From: Ted Dennison @ 2000-05-30 0:00 UTC (permalink / raw) In article <8h0rne$h92$1@nnrp1.deja.com>, Robert Dewar <robert_dewar@my-deja.com> wrote: > In article <8h0qd0$g1q$1@nnrp1.deja.com>, > So unless you have a very definite statement, preferably in > writing signed, an email message is unlikely to be taken as > a formal contract, or at least authenticity questions can > easily be raised in this context, you may be in trouble. > The statement must CLEARLY give the author's permission > for incorporating the code under whatever licence you are > using. I can see where you are comming from here. But I'm afraid that requiring a formal written and notarized contract for every little submission would have a very chilling effect on such a small project. I know for a fact that I don't really have the money to develop such a contract. Note that my main co-contributor so far (Christ Grein) is based in Germany. How do I know what the legal ramifications of that are? Perhaps there's some law there that doesn't allow people to sign away copyright intrests in the way my homespun contract requires. It would take a true international IP laywer to sort that out, and I just can't afford to do that. > There is a lot of sloppiness in these kind of procedures with > code wandering around, but that sloppiness can come back to > haunt you in court later on. True. But when you are talking total hobbyist/volunteer efforts, I'm afraid a certian amount of such legal "slop" is going to be unavoidable. If I do something email-based that makes it clear what is going to happen to a submission, and that the submitting person agrees to that, that ought to get me about 90% of the way there, without turning off prospective submittors too much. It should at least prevent misunderstandings, like the one in your scenario. It would be absolutely no defense against a company with deep pockets and a commitment to breaking the license. But at my present financial level, no amount of formal written documents would save me in that sceario either. -- T.E.D. http://www.telepath.com/~dennison/Ted/TED.html Sent via Deja.com http://www.deja.com/ Before you buy. ^ permalink raw reply [flat|nested] 39+ messages in thread
end of thread, other threads:[~2000-06-06 0:00 UTC | newest] Thread overview: 39+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2000-05-27 0:00 "proprietary", was Re: ada on linux tmoran 2000-05-28 0:00 ` Robert Dewar 2000-05-28 0:00 ` Ken Garlington 2000-05-28 0:00 ` Robert Dewar 2000-05-30 0:00 ` Ted Dennison 2000-05-30 0:00 ` Ken Garlington 2000-05-28 0:00 ` tmoran 2000-05-28 0:00 ` David Starner 2000-05-28 0:00 ` Robert Dewar 2000-05-28 0:00 ` tmoran 2000-05-28 0:00 ` David Starner 2000-05-29 0:00 ` Robert Dewar 2000-05-29 0:00 ` tmoran 2000-05-29 0:00 ` Robert Dewar 2000-05-30 0:00 ` Ted Dennison 2000-05-30 0:00 ` Robert Dewar 2000-05-30 0:00 ` bill 2000-05-31 0:00 ` Florian Weimer 2000-06-01 0:00 ` Geoff Bull 2000-06-03 0:00 ` Robert Dewar 2000-06-03 0:00 ` Robert Dewar 2000-06-03 0:00 ` tmoran 2000-06-03 0:00 ` Dale Stanbrough 2000-06-03 0:00 ` Jeff Creem 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` Jeff Creem 2000-06-06 0:00 ` GPL distribution rules (was: "proprietary") Larry Kilgallen 2000-06-05 0:00 ` "proprietary", was Re: ada on linux Robert Dewar 2000-06-05 0:00 ` Geoff Bull 2000-06-05 0:00 ` tmoran 2000-06-05 0:00 ` Geoff Bull 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` Robert Dewar 2000-06-05 0:00 ` tmoran 2000-05-30 0:00 ` Ted Dennison 2000-05-30 0:00 ` Robert Dewar 2000-05-30 0:00 ` About AdaOS Didier Utheza [not found] ` <WCBZ4.4122$XX4.63232@news-east.usenetserver.com> 2000-06-01 0:00 ` Didier Utheza 2000-05-30 0:00 ` "proprietary", was Re: ada on linux Ted Dennison
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox