Re: [racket-users] Licence guidance
On Thu, Sep 27, 2018 at 01:21:15PM +0100, Norman Gray wrote: > > Greetings. > > On 27 Sep 2018, at 3:48, Anthony Carrico wrote: > > > On 09/26/2018 05:32 PM, Deren Dohoda wrote: > > > > > I put a package up but it has no license info in the code. I would > > > add > > > one which is the most permissive possible that wouldn't cause > > > conflict. > > > I guess this is BSD? MIT? > > > > > > In this case, don't license your code, declare it to be in the public > > domain. > > That doesn't necessarily solve the problem, or at least not internationally. > > In UK law, for example, 'public domain' means simply 'known to the public', > and doesn't have a link to licence information. Also, it seems that there > isn't the notion of 'unowned (intellectual) property', so that 'I place this > in the public domain' could at most be interpreted as a vague disavowal of > interests. That is, it would be an absence of a statement of a licence, > rather than a statement of an absence of a licence. > > Thus the BSD licence is probably the most permissive thing that's still > unequivocally recognisable as a licence. One of the Creative Commons licences is equivalent to the concept of public domain we eem to want, formulated as a licence for jurisdictions where "public domain" isn't a concept. -- hendrik -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
Hi, Take a look at MPLv2, it's a share-and-share-alike but at file level so you can include it in proprietary code if you want. This is a good license for hackers and businesses. kr/sjm On Mon, 24 Sep 2018, 18:04 Stephen De Gabrielle, wrote: > c) anything else? > -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
Greetings. On 27 Sep 2018, at 3:48, Anthony Carrico wrote: On 09/26/2018 05:32 PM, Deren Dohoda wrote: I put a package up but it has no license info in the code. I would add one which is the most permissive possible that wouldn't cause conflict. I guess this is BSD? MIT? In this case, don't license your code, declare it to be in the public domain. That doesn't necessarily solve the problem, or at least not internationally. In UK law, for example, 'public domain' means simply 'known to the public', and doesn't have a link to licence information. Also, it seems that there isn't the notion of 'unowned (intellectual) property', so that 'I place this in the public domain' could at most be interpreted as a vague disavowal of interests. That is, it would be an absence of a statement of a licence, rather than a statement of an absence of a licence. Thus the BSD licence is probably the most permissive thing that's still unequivocally recognisable as a licence. Best wishes, Norman -- Norman Gray : https://nxg.me.uk -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
On 09/26/2018 05:32 PM, Deren Dohoda wrote: > I put a package up but it has no license info in the code. I would add > one which is the most permissive possible that wouldn't cause conflict. > I guess this is BSD? MIT? In this case, don't license your code, declare it to be in the public domain. -- Anthony Carrico -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
If you want a permissive license, the FSF itself says that "Apache 2.0 is best" because it addresses patent issues: https://www.gnu.org/licenses/license-recommendations.html#small -Philip On Wed, Sep 26, 2018 at 4:32 PM Deren Dohoda wrote: > I put a package up but it has no license info in the code. I would add one > which is the most permissive possible that wouldn't cause conflict. I guess > this is BSD? MIT? > > On Tue, Sep 25, 2018, 12:30 PM Neil Van Dyke wrote: > >> BTW, I don't know the status of possible new GPL and LGPL versions in >> progress, but, if any Racket people have some insights into how to >> improve the "linking" concepts or some other aspect, in the spirit of >> FSF goals, the FSF has seemed open to comments. If you don't know who >> else to contact, you can always email RMS. >> >> You can get as technical as necessary. Organizationally, besides the >> FSF being founded by and attracting some high-powered techies, some >> advisors are very noteworthy Scheme people. Also, a Scheme with >> multiple `#lang`s (though it wasn't called `#lang`) was actually a >> central part of the GNU roadmap, from early on. (And probably would've >> been in popular use atop Linux for over a decade now, but for another >> party's business decision, outside of FSF's control, not a technical or >> acceptance reason.) >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Racket Users" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to racket-users+unsubscr...@googlegroups.com. >> For more options, visit https://groups.google.com/d/optout. >> > -- > You received this message because you are subscribed to the Google Groups > "Racket Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to racket-users+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
I put a package up but it has no license info in the code. I would add one which is the most permissive possible that wouldn't cause conflict. I guess this is BSD? MIT? On Tue, Sep 25, 2018, 12:30 PM Neil Van Dyke wrote: > BTW, I don't know the status of possible new GPL and LGPL versions in > progress, but, if any Racket people have some insights into how to > improve the "linking" concepts or some other aspect, in the spirit of > FSF goals, the FSF has seemed open to comments. If you don't know who > else to contact, you can always email RMS. > > You can get as technical as necessary. Organizationally, besides the > FSF being founded by and attracting some high-powered techies, some > advisors are very noteworthy Scheme people. Also, a Scheme with > multiple `#lang`s (though it wasn't called `#lang`) was actually a > central part of the GNU roadmap, from early on. (And probably would've > been in popular use atop Linux for over a decade now, but for another > party's business decision, outside of FSF's control, not a technical or > acceptance reason.) > > -- > You received this message because you are subscribed to the Google Groups > "Racket Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to racket-users+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
BTW, I don't know the status of possible new GPL and LGPL versions in progress, but, if any Racket people have some insights into how to improve the "linking" concepts or some other aspect, in the spirit of FSF goals, the FSF has seemed open to comments. If you don't know who else to contact, you can always email RMS. You can get as technical as necessary. Organizationally, besides the FSF being founded by and attracting some high-powered techies, some advisors are very noteworthy Scheme people. Also, a Scheme with multiple `#lang`s (though it wasn't called `#lang`) was actually a central part of the GNU roadmap, from early on. (And probably would've been in popular use atop Linux for over a decade now, but for another party's business decision, outside of FSF's control, not a technical or acceptance reason.) -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
Thank you all. Sounds like a case-by-case approach is probably best. I did find the github licence advice pages. It seems that not choosing a licence is probably a bad choice when publishing a racket package: https://choosealicense.com/no-permission/ I can't believe I missed the Racket licence pages! https://download.racket-lang.org/license.html and https://docs.racket-lang.org/license/index.html Thanks again, Stephen On Mon, Sep 24, 2018 at 8:55 PM Neil Van Dyke wrote: > Yes, it could be that LGPL is not the best for Racket package authors > who intend something analogous to LGPL for C libraries. (Or who intend > not necessarily that, but something in the neighborhood of that flavor > or degree.) > > Law quickly gets way outside my expertise, and the finer points seem to > need the most masochistic brainiac lawyers to wrestle, so I'd only like > to throw out 3 quick comments at this time: > > * The Racket package system, and ways that systems using Racket are > distributed in practice, as well as the market potential, have changed > since I picked a license. > > * I think industry and open source practice, the information technology > infrastructure, and the world have all changed substantially since open > source people really-really rethought licenses. > > * Having been engaged with GNU and FSF for many years, and occasionally > talking with RMS, including about "linking" subtleties and > implications... the various FSF technicalities are usually imperfect, or > not entirely clear. (The reason seems to be difficulty of legal > constructs that capture intent over decades, for evolving technology, > and without the benefit/curse of a legislative system. Though the FSF > remains an interactive component, for one-on-one interpretation, as well > as leaving the "or any later version" evolutionary hook in the language.) > > -- > You received this message because you are subscribed to the Google Groups > "Racket Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to racket-users+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
Yes, it could be that LGPL is not the best for Racket package authors who intend something analogous to LGPL for C libraries. (Or who intend not necessarily that, but something in the neighborhood of that flavor or degree.) Law quickly gets way outside my expertise, and the finer points seem to need the most masochistic brainiac lawyers to wrestle, so I'd only like to throw out 3 quick comments at this time: * The Racket package system, and ways that systems using Racket are distributed in practice, as well as the market potential, have changed since I picked a license. * I think industry and open source practice, the information technology infrastructure, and the world have all changed substantially since open source people really-really rethought licenses. * Having been engaged with GNU and FSF for many years, and occasionally talking with RMS, including about "linking" subtleties and implications... the various FSF technicalities are usually imperfect, or not entirely clear. (The reason seems to be difficulty of legal constructs that capture intent over decades, for evolving technology, and without the benefit/curse of a legislative system. Though the FSF remains an interactive component, for one-on-one interpretation, as well as leaving the "or any later version" evolutionary hook in the language.) -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
David linked above to https://download.racket-lang.org/license.html, where the Racket maintainers (who are not lawyers, and neither am I) explain their interpretation of how "linking" in the LGPL applies to Racket. I think it's worth copying here for the record: Since the LGPL license that Racket uses was originally designed for C > programs, parts of it require some interpretation to apply to Racket in > detail. The following is how the Racket maintainers interpret the license. > > First, if you distribute your Racket application in source form or as > compiled bytecode files, the Racket license does not restrict you at all. > > Second, if you distribute your Racket application as compiled binary > generated by raco exe, there are no requirements placed on the licensing of > your software. However, the LGPL requires that you make it possible to > re-link your software with modified versions of Racket. This means, > basically, that you need to provide the compiled bytecode files used to > produce the compiled binary, if requested by someone who got your software > from you. Note that this does not mean that your software has to be made > open source, nor do you have to give the source code to anyone, nor do you > have to make the compiled bytecode files available to the public or let > other people redistribute them. Furthermore, this is not revealing any more > of your source code than the raco exe format, since the bytecode is > embedded in an extractable way in the resulting executable. > I would add the observation that the word "easily" is subjective, and doesn't appear in the LGPL anyway, but I don't think this process for Racket is objectively more difficult than replacing a library for a C program. The process for C might be more familiar, but that's only because Racket is still in the process of conquering the world :) -Philip On Mon, Sep 24, 2018 at 2:18 PM wrote: > Are you sure that's a wise choice of license? > > Racket does not dynamically link to Racket libraries when applications are > deployed as compiled executables - as far as I can see, the standard module > system does not link dynamically in the sense required by the LGPL(*). > Therefore, LGPL doesn't allow anyone to distribute his or her Racket > application as compiled executable without making the source code available > on request, too, whenever that source code was made with an LGPL Racket > library. So for users of your library, LGPL is pretty much equivalent to > GPL. They have to provide the source code, or the program has to load the > library with dynamic require at runtime, if that's possible at all. > > (*)=the user must be able to easily update the library that the executable > uses, e.g. by linking to another dynamic library > > - Ursprüngliche Nachricht - > Von: > "Neil Van Dyke" > > An: > "Stephen De Gabrielle" , "Racket Users" < > us...@racket-lang.org> > Cc: > > Gesendet: > Mon, 24 Sep 2018 13:06:15 -0400 > Betreff: > Re: [racket-users] Licence guidance > > > What's seemed to work over many years for my Racket open source packages > (a couple of which are useful things that would be expensive to rewrite) > is LGPL (initially version 2.x, but lately version 3), plus a statement > to contact me about other possible licenses. > > My thinking was, LGPL suggests the idea of sharing in the same spirit, > but has very modest limitations that aren't a barrier to most commercial > use. At the same time, if someone wanted to make commercial > inconsistent with even those modest limitations (e.g., make the package > a closed network service), that was OK, but I'd want some of that money, > in lieu of common benefit warm-fuzzies. > > I don't recall ever being contacted about a different license other than > LGPL. Some of my Racket packages are used by an important large > closed-source production server system, for example, without any > different license. (Though, at one point, someone who wanted to > incorporate my CSV library into an open source something initially > seemed to be interested in an older version that was LGPLv2 rather than > v3, perhaps for compatibility with their license at the time, but then > they quickly straightened that out somehow.) > > That said, I don't know whether LGPLv3 is the best default license for > third-party Racket packages. Maybe that should be revisited. > > Today, Racket hasn't yet taken off commercially nearly as much as it > might've,[1] and maybe the biggest concern with the licenses of open > source Racket packages shared altruistically is that we not > inadvertently lock out some positive later effort (because, e.g., we > didn't leave a permissive enough license before we went
Re: [racket-users] Licence guidance
Are you sure that's a wise choice of license? Racket does not dynamically link to Racket libraries when applications are deployed as compiled executables - as far as I can see, the standard module system does not link dynamically in the sense required by the LGPL(*). Therefore, LGPL doesn't allow anyone to distribute his or her Racket application as compiled executable without making the source code available on request, too, whenever that source code was made with an LGPL Racket library. So for users of your library, LGPL is pretty much equivalent to GPL. They have to provide the source code, or the program has to load the library with dynamic require at runtime, if that's possible at all. (*)=the user must be able to easily update the library that the executable uses, e.g. by linking to another dynamic library - Ursprüngliche Nachricht - Von: "Neil Van Dyke" An:"Stephen De Gabrielle" , "Racket Users" Cc: Gesendet:Mon, 24 Sep 2018 13:06:15 -0400 Betreff:Re: [racket-users] Licence guidance What's seemed to work over many years for my Racket open source packages (a couple of which are useful things that would be expensive to rewrite) is LGPL (initially version 2.x, but lately version 3), plus a statement to contact me about other possible licenses. My thinking was, LGPL suggests the idea of sharing in the same spirit, but has very modest limitations that aren't a barrier to most commercial use. At the same time, if someone wanted to make commercial inconsistent with even those modest limitations (e.g., make the package a closed network service), that was OK, but I'd want some of that money, in lieu of common benefit warm-fuzzies. I don't recall ever being contacted about a different license other than LGPL. Some of my Racket packages are used by an important large closed-source production server system, for example, without any different license. (Though, at one point, someone who wanted to incorporate my CSV library into an open source something initially seemed to be interested in an older version that was LGPLv2 rather than v3, perhaps for compatibility with their license at the time, but then they quickly straightened that out somehow.) That said, I don't know whether LGPLv3 is the best default license for third-party Racket packages. Maybe that should be revisited. Today, Racket hasn't yet taken off commercially nearly as much as it might've,[1] and maybe the biggest concern with the licenses of open source Racket packages shared altruistically is that we not inadvertently lock out some positive later effort (because, e.g., we didn't leave a permissive enough license before we went to live off-grid on a sunny beach with our Racketeering plunder). I'm not saying that licenses shouldn't be totally permissive, though, or we'd just release everything to the public domain. Racket is pretty neat, and, for example, there's still an icky element or two in industry that have been known to sabotage projects, and licenses can offer some defense. I think there's little danger of Racket ever being a threat or underhanded opportunity to anyone -- but, with all the students around, we should try to serve example of good open source licensing, in the best spirit of engineering and goodwill. [1] It's not too late. :) https://www.neilvandyke.org/racket-money/ -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
What's seemed to work over many years for my Racket open source packages (a couple of which are useful things that would be expensive to rewrite) is LGPL (initially version 2.x, but lately version 3), plus a statement to contact me about other possible licenses. My thinking was, LGPL suggests the idea of sharing in the same spirit, but has very modest limitations that aren't a barrier to most commercial use. At the same time, if someone wanted to make commercial inconsistent with even those modest limitations (e.g., make the package a closed network service), that was OK, but I'd want some of that money, in lieu of common benefit warm-fuzzies. I don't recall ever being contacted about a different license other than LGPL. Some of my Racket packages are used by an important large closed-source production server system, for example, without any different license. (Though, at one point, someone who wanted to incorporate my CSV library into an open source something initially seemed to be interested in an older version that was LGPLv2 rather than v3, perhaps for compatibility with their license at the time, but then they quickly straightened that out somehow.) That said, I don't know whether LGPLv3 is the best default license for third-party Racket packages. Maybe that should be revisited. Today, Racket hasn't yet taken off commercially nearly as much as it might've,[1] and maybe the biggest concern with the licenses of open source Racket packages shared altruistically is that we not inadvertently lock out some positive later effort (because, e.g., we didn't leave a permissive enough license before we went to live off-grid on a sunny beach with our Racketeering plunder). I'm not saying that licenses shouldn't be totally permissive, though, or we'd just release everything to the public domain. Racket is pretty neat, and, for example, there's still an icky element or two in industry that have been known to sabotage projects, and licenses can offer some defense. I think there's little danger of Racket ever being a threat or underhanded opportunity to anyone -- but, with all the students around, we should try to serve example of good open source licensing, in the best spirit of engineering and goodwill. [1] It's not too late. :) https://www.neilvandyke.org/racket-money/ -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [racket-users] Licence guidance
On Mon, Sep 24, 2018 at 6:03 AM, Stephen De Gabrielle < spdegabrie...@gmail.com> wrote: > Hi, > > I sometimes see Racket packages on PLaneT or Github, but lack a licence. > > I don’t feel I can redistribute or fork abandoned code if it lacks a > licence. (I can give an example of an 11yo abandoned project that I’d love > to fork but can’t because it lacks a licence) > > With that in mind- what licences should be used when making code available > on the package repository/github in the following situations: > > a) general purpose library that I am happy for the broader community of > evelopers to use without restraint - but is unlikely to ever meet to be > included in Racket e.g. I have an number checksum validator (UK NHS ID > number) > - is dual licence Apache/MIT appropriate? or is it completely up to me. > > b) library, tool or DrRacket plugin that may(or may not) become part of > the Racket distribution > - is dual licence Apache/MIT appropriate or should it also be LGPL? > > c) anything else? > > Kind regards, > > Stephen > It depends very much on what you want to accomplish. If your goal is to help people get things done in the general case then you should choose a liberal license such as BSD or MIT. These will guarantee that you get credit for your work and that the person using that work can make a living off your code or enable someone else (including a company) to make a living off your code. If your goal is to ensure that all users have maximal freedom with their software and you're okay with undermining the system of intellectual property that enables people to make money directly off of selling software (as opposed to indirectly, such as by consulting on open source software), then you should choose a restrictive license such as GPL. If you want to be somewhere in the middle, you could choose a license such as LGPL, which allows people to use your code as part of something from which they make a living, but not to directly make a living off your code. As to things that will be included under DrRacket, the dev team can speak to that better than I, but Racket itself is distributed under an interpretation of the LGPL: https://download.racket-lang.org/license.html Dave > > -- > Kind regards, > Stephen > -- > Ealing (London), UK > > -- > You received this message because you are subscribed to the Google Groups > "Racket Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to racket-users+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
