That’s the 0BSD. I called it the Free Public License when submitting to OSI, but we changed it to 0BSD because Landley had apparently come up with the same license a few years before me.
Thanks for this. 0BSD has been my default license since I discovered it. My only issue with the MIT license was the requirement for attribution, which always seemed like an unnecessary pain. 0BSD fixes that.
It’s the same text, which is just the ISC license sans the “ provided that the above copyright notice and this permission notice appear in all copies” clause.
The only differences:
- I didn’t use the “Copyright Boaty McBoatface 2022” line, but that’s not a big enough change to justify having two licenses.
- The license was approved as the FPL-1.0.0, which I believe was the first OSI-approved license that user semantic versioning. This isn’t necessary, but I was mildly excited about it.
- The 0BSD doesn’t actually inherit from any other BSD license, the “BSD” bit is more of a spiritual nod that I found more confusing than helpful.
0BSD is actually incompatible with a desired pure CC0. Unlike CC0, 0BSD keeps copyright intact. It's just a license with no conditions except the usual disclaimer; CC0 attempts to cause a public domain dedication wherever possible. Consider the following:
1. Let there be a file F. F is released to the public under 0BSD.
2. Let there be a project P. P includes code from F. The license notice of F is scrubbed as part of the inclusion process.
3. P is released as CC0.
The author of P cannot release the project as CC0 because the author has no right dedicate another person's code (F) to the public domain. Because 0BSD doesn't cause a public domain dedication in jurisdictions that accept doing so, F would have to continue being separate and explicitly exempt from the effects of CC0 to be an accurate of its usage terms. Effectively, this makes 0BSD be a de-facto attribution license in a CC0 project. In practice, it makes no difference, but it's legally deeply inelegant.
...isn't that literally the opposite of public domain? You can't have it both be public domain with no copyright and ALSO use copyright to enforce what people do with their copies of it.
CC0 isn't an OSI-approved license. (It was withdrawn from consideration around patent language concerns.) MIT-0 (or probably 0BSD now as well) is probably the closest you get to a public domain license to the degree that putting something in the public domain is recognized in a given jurisdiction.
It lacks OSI approval, though -- which is a big one. I suppose someone could jump on the OSI mailing lists to get it through the approval process, but it hasn't been done yet.
All of us have direct experience with both OSI and FSF license processes. We have no desire to deal with them again. Even for a thoroughly noncontroversial permissive license.
> All of us have direct experience with both OSI and FSF license processes. We have no desire to deal with them again.
As a more casual person looking to use a license, this is basically the opposite of what I want to read.
As a non-lawyer, I have little chance of vetting a license myself. As a person with life experience, I have full appreciation for the delicacy of these kinds of licenses and legal issues.
If a license has consensus from OSI, FSF, Debian, maybe a FANG company or two, etc., then I have some trust that a kind of validation and peer review has occurred. I fear that using a license for which this has not occurred puts the users of my software at risk, or at least shifts the burden of future validation on to them.
TLDR: everybody knows what MIT, BSD, Apache, and the various GPL licenses imply. I think other bespoke licenses have a pretty high bar to clear before they become truly useful.
I've written a bit on my experiences with OSI review in particular, and also the list of licenses it has approved. See, e.g., https://writing.kemitchell.com/2019/05/05/Rely-on-OSI.html. In a nutshell: people think there's a team of lawyers back there, counting angels on the heads of fine legal pins. But it's actually a highly politicized mailman list where activist-types do most of the talking.
I put energy into cofounding Blue Oak Council to publish rigorous resources, like our permissive license list, https://blueoakcouncil.org/list, in large part out of disappointment and dread, pulling back the curtain on the institutional processes for license review. As lawyers, we need functional resources like that license list, to incorporate by reference into contracts and policies. Fundamentally political artifacts, like OSI or FSF or Debian whitelists, are type errors in those contexts.
> everybody knows what MIT, BSD, Apache, and the various GPL licenses imply
I'm afraid that's not true. I would say there is broad agreement on many of the core aspects. And in many situations, practically, the specific license terms don't matter nearly as much as widely help expectations. But key concepts affecting fairly common situations remain unclear, and those details come out when there's money or strategic leverage on the line. See, e.g. the list Heather Meeker keeps at https://heathermeeker.com/open-source-faq/what-are-the-most-... See also the whole debate on permissive licenses and standards-essential patents: https://writing.kemitchell.com/2019/10/03/Open-Standards.htm....
Thanks for chipping in on this thread, you’re expressing things nice and clearly, now I have a few more links for describing this unhappy situation in open-source licensing.
Any chance you could change the slug on that first link from Rely-on-OSI to Dont-Rely-on-OSI-Approval or similar? The current URL suggests the opposite of the article’s content. Hopefully adding a redirect isn’t too onerous.
> You must ensure that everyone who gets a copy of any part of this software from you, with or without changes, also gets the text of this license or a link to https://blueoakcouncil.org/license/1.0.0.
To me this sounds like a viral public domain, so in a sense less permissive than MIT & co. If I modify the work but still have to put the text of this license when publishing it, the modified/derivative will be licensed under this license as well. Is this a fair assessment or am I missing something? I don't see how someone could take the work and use it in a proprietary software for example.
> If I modify the work but still have to put the text of this license when publishing it, the modified/derivative will be licensed under this license as well.
How does that follow? Compare:
MIT:
> The above copyright notice and this permission notice (including the next paragraph) shall be included in all copies or substantial portions of the Software.
BSD:
> 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
>
> 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
I'm not sure the reason for bringing up that clause, when the same clause exists in the MIT license:
"Permission is hereby granted [to do so-and-so] subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software."
If you're shipping MIT-licensed software (or anything derived from MIT-licensed software) without also shipping a copy of the license, then you're not in compliance with the license.
The problem with CC0 is it's very wordy and very legaleeze, which means I don't want to bother trying to understand it, and potential users won't either.
A one-line sentence should be more than enough to clarify my intent.
Public domain. Concise. Effective. Choose any two.
The issue is that donating to the “public domain” isn’t a thing in many jurisdictions, so if you want a license that is effective everywhere then it needs to be wordy. If you are happy to retain copyright and other rights and simply want to grant other people a license to use you work then just use MIT.
I hear this a lot, but what jurisdictions and why? On the face of it, it seems batshit insane to me that there would be a place where I can’t waive my property rights on something.
> With regard to the transferability and the waiving of the copyright the German copyright law can be considered as one of the strictest systems in the world. Main reason is the strict monistic approach the German copyright law bases on. Key feature of this approach is the concept that, in principle, the copyright/author’s right itself can neither be transferred to another person nor
waived by the author herself. The German author’s right consists of two parts, the moral rights
and the exploitation rights. The moral rights are – as a rule – personal rights that are bound to the person of the creator (or, after her death, her legal heirs), i.e. they can neither be transferred nor waived. Since moral and exploitation rights are considered as inseparable parts of the author’s
right as a whole (monistic approach) the exploitation rights cannot – in principle – transferred or waived by contract as well. However it is naturally possible to license the use of the work i.e. to transfer rights to use a protected work even on a large scale. Such licenses can practically lead
(nearly) to the same result as an assignment or waiver of rights.
Germany would be one example. Public Domain exists, but only as the state something enters after copyright expired naturally 70 years after the death of the author. You can't release things into public domain.
As for why: nobody considered that case when writing our copyright laws, and nobody bothered to change it. Copyright as designed can't be transferred (except through inheritance), to avoid exploitation of the original creator. As a consequence you can't really get rid of it, you can only grant licenses.
It's hardly the only right you can't get rid of, and CC0 tries to deal with some of them, like the right to one's one image which in the shortest possible form says that you can't create or publish a picture of a person without their consent (but as you can imagine is way more complicated than that). You could say that the German legal framework isn't about maximizing freedom, it's about maximizing happiness, and sometimes being able to give away a right or freedom will on balance cause more harm than good.
> Public domain. Concise. Effective. Choose any two.
A workaround to the "concise" problem would be a commonly understood convention for indicating that a file is CC0 with a single line comment. The officially recommended boilerplate for CC0 is painfully cumbersome:
// Available under http://creativecommons.org/publicdomain/zero/1.0
Come to think of it, it sure would be nice if "Available under [URL]" came to be a convention for FOSS licensing in general. I'm choosing "available" rather than "licensed" because it seems as though it would be generalizable to both copyright licenses and public domain dedications. (Caveat: though I'm fairly familiar with FOSS licensing for a layperson, I'm not a lawyer and so this would need to be reviewed by a lawyer to ensure for drafting soundness.)
I try to look this up and there isn't a concrete answer. There are countries with no or unclear Copyright terms, but I cant find countries with no public domain.
One example is in Germany you can't give up all your rights (the sticky rights are often called "moral rights"). In that case you want to fallback to a permissive license.
There are countries (eg. France) where moral rights are inalienable.
There are countries where you cannot put something into the public domain (Germany). Even countries where there isn't such a thing as the public domain.
You end up needing to write something which is clearly a license, yet acts like the work is in the public domain, and make it watertight (nearly) worldwide. So you end up with CC0.
Luckily you don't need to write anything very wordy to use CC0. It's actually pretty simple and short at your end. Creative Commons have already done the hard bit.
No, no, no, no. Nope. That's enough in US but never enough (like, legally impossible) in Europe or parts of Asia (that's why CC0 has a fallback permissive license).
I'm curious... if Creative Commons can't even create a public domain license that applies to both US and Europe, then what would it actually take to do so? Seems insane that any system can't accept "this is public domain" as all you need.
> Seems insane that any system can't accept "this is public domain" as all you need.
Except that's exactly (legally, anyways) sane in Europe (not EU, it predates EU). Read about moral rights, and you will learn the complexities of what the US squishes into a single copyright regime.
Note: moral rights was translated from French legal term droits moraux, there's no religious meaning in moral here.
This is up to debate. I am a european resident and I think it is insane, that the reason I am not free to donate my work into the public domain is supposed to be "moral rights".
It's not that anyone is denying you your right. You aren't "free" to donate your work to flutbark bargleboff either not because its forbidden but because its undefined.
Contract your lawmaker and have them define it legally. Its undefined because there isn't money to be gained by giving stuff away and for practical purposes current solutions seem to be work good enough for practical purposes for those who want to do so robs any such movement of much momentum thus far.
Of course they are. I am not free do do with my IP as I want. (gifting it to everyone).
I have to keep ownership. (even though I reject the concept of IP alltogether)
But you are right with your explanation why it is like this. There is no money in it to change it for good and current solutions work "good enough" for practical purposes (giving everyone the right to use my IP).
So I will likely also just go on use the existing solutions, instead of running against walls and trying to change the laws.
What is the status of work that no one knows where it came from?
A recipe for a cake is written on a paper and left on a bulletin board in town. It has no name and no one witnessed it being posted.
Everyone sees it and now knows how to make this cake.
The author may or may not be dead for 70 years.
Is it illegal to even document this cake recipe? You couldn't legally recount it in a news article or history text book without being able say who the author was and prove that they either granted a copy right or are known to be dead for 70 years?
Or worse, you are not only incentivised but essentially obligated to claim the authorship yourself just so you have someone to assign it to?
The thing undeniably exists regardless that the legal structure provides no place for it.
It's patently ridiculous. (guilty, terrible pun intended).
That's called an orphan work. You can apply §61 UrhG, which in this case would mean donating it to an archive, museum, library or university, who can then search for the author and after turning up empty are allowed to publish it. From there you can probably quote it, citing them as your source.
And there we come to the nuance of copyrights, at least in the US: you can't copy the text of the recipe, but the recipe itself don't have protection itself. Specifically, the text of the recipe is literature (which is protected), the recipe itself are instructions (which is not protected). This is the whole justification into clean-room design and reverse-engineering in software.
Also psuedonymous or even anonymous publications do have copyright (insert a legalheap of asterisks here). This actually sadly results into something called "orphan works", where they don't knew (or if they knew cannot find the next-of-kin/successor company/maybe even the government where the company was incorporated) where they can ask for permission.
I only imagine something like "but you give up the right to sue me if any harm comes from whatever you do with this" would need to be appended to "you can do anything with this".
No. There are certain rights to a work you cannot waive, no matter how much you want to. Even less without actually specifying what "anything" entails.
While you can not waive certain rights, you absolutely can proclaim "you can do anything with this". You can choose to not exercise your legal right to attribution, for example. So what you’re saying is not in conflict with the person you’re replying to.
(You can argue that there’s some legal risk at hand here; for example, if someone writes a book with the above and you start selling copies of it claiming you wrote it, the author can still come back and demand that you credit them as author and have that right enforced in court)
I think this is great. Case in point: blanket EULA for an image sharing site can not remove your right to be attributed, should your photo end up in an ad campaign or a newspaper. This means that many European newspapers actually make sure to not misattribute photographs, which has the side-effect of improving transparency in news reporting. So grabbing a photo online and wrongfully claiming it’s from the wrong time, place and source is not something that tends to happen in practice. And if it does, there’s a clear path to recourse that doesn’t require "lawyering up", you just file a claim with evidence.
1) Hipp, Wyrick & Company, Inc. (Hwaci, the legal owner of the code) does not actively enforce its rights available. This does not mean it will not affect operations in certain countries, while computer code is not (yet) affected, music definitely is (see GEMA in Germany). Additionally, in an adverse purchase of a company Hwaci could be bought by a scrupulous corporation (à la patent trolls of today) and enforce the copyright in countries where Hwaci has residual rights. If somehow that did not happen but Hwaci faces bankruptcy and wound up in favour to the State of North Carolina (the default option), we will have a similar problem as a particular Canadian company that unfortunately currently escapes my mind. This is what worries large companies: a permissive license which is recognised internationally cannot be annulled (unless they breached the very simple requirement), sure for new users it would suck (even GPL doesn't close the hole), but the new company cannot sue companies with existing licences because they have a very permissive license.
I've long been aware of their $6k warranty of title. Only someone from Hwaci could give an accurate count of purchasers (they only say "Most people use SQLite without any kind of license or support"), but I would bet that not even half of the developer-users and distributors of sqlite directly (let alone those of sqlite indirectly through other products or libraries that bring it in or embed it, which they are under no obligation to disclose) in countries like Germany have actually purchased one. I doubt it's even a significant money-maker for them given their range of available things for purchase (https://www.sqlite.org/prosupport.html). It's not a problem in practice, it's exceedingly unlikely to ever become a problem, but sure, a lawyer can find a vulnerability in everything, even with GPL (which is part of why some US companies have a blanket "no GPL" policy).
I'm glad they offer the warranty of title and take money from companies with edgy lawyers, though. $6k is small as legal fees go. It's in general a lazy monetization strategy I broadly agree with and I've been hoping GitHub under MS's stewardship would help spearhead a simple way for people/companies to buy secondary-licensed copies of software without having to contact the author about it. (I think I got the idea from Zed Shaw around the time MS bought GH.) Like, a lot of software out there isn't a mess of contributors like the linux kernel each holding their own copyright, and thus much of it can be relicensed or dual-licensed without fuss. When people license something under the AGPL or even public domain if they're copying sqlite, sometimes they include something in the readme about being open to selling an alternatively licensed version/warranty of title to companies who have some issue. But even if they don't explicitly mention it, companies can still fruitfully contact such authors and make a deal. This happens all the time on twitter of all places, with journalists asking some user (who may not even be legally identifiable from just their handle) to re-use their apparently original photo or video. Sometimes the user just gives permission, sometimes they demand "pay me", sometimes a payment is actually made. Douglas Crockford has given permission (not sure if he got any money out of it) at least to IBM to use his software for evil, again showing his infamous clause is only a problem for people who decide to make it one.
Basically, I'd rather not tell people how they can license (or not license, as per public domain declarations) their code, and while it's fine to be aware of usability tradeoffs and it's fine to know that some edgy companies or users in other countries might write you off for your choice, it's important not to overstate the actual risks involved relative to the base rate risks already out there (like patent trolls) and it's important to recognize that solving the problems in ways even the edgiest of lawyers can be mostly happy with is often possible. Having blanket policies instead of pursuing those solutions is also a choice people are free to make.
(Writing this comment led me to https://web.archive.org/web/20120510151444/http://methodlogi... which is interesting in that it could suggest how further knowledge of tradeoffs and future pushback in the beginning may have led sqlite to be put under a license, however I don't suppose anything legally stops the authors agreeing to relicense in a new version since they retain a clean title, it's just there's no compelling practical reason like seeking more contributors.)
Neither, really, it’s just that "public domain" is a US-specific legal term that doesn’t have equivalents in most countries.
For one example, in Sweden you can never fully sign away the right to have creative work attributed to your name (until 50/70 years after creation/creator death, depending on nature of the work) as a creator. You can make a pinky promise to not claim that right, but you can’t fully waive it so you could always have a change of heart and demand to be credited by name later. IIRC this came from a notion of protecting artists and photographers from predatory contracts from the print industry or something similar.
Commercial rights etc can all be fully transferred, of course.
Yes. At least if you want people who have to talk to lawyers to use the code. Lawyers don't like that. For some anecdotes see this talk https://youtube.com/watch?v=MkJkyMuBm3g (worth the hour, althoug Rob Landley write the same thing on his website too but to lazy to Google that.
It's undesired by people who want to exploit other people's work commercially or fantasize that they somehow they might want to do so in the future. Kind of like poor people who don't want to tax the rich because they fantasize that they may someday be rich.
Google et. al (and OSI for that matter) will find no such alike-license acceptable until a platform using that license becomes too valuable to ignore, like Linux with GPLv2. So the calculus has to tilt such that giving their work away for free is worth it because they get so, so much work for free. And even then they will try to work around it like with Fuschia.
Plus since everything is served over the internet these days unless a license includes AGPL-type clauses the alike requirement will not attach.
I was running `npm init` recently on a new private project, I meant to write UNLICENSED which means "all rights reserved", but I accidentally missed the "d" at end and it got marked with "Unlicense" license.
Personally it's a mix of principle and practicality. On principle I'd like to be able to put code up on Github, tell people they can use it however they like and to credit me if it suits them, but otherwise I don't want the copyright or any other associated responsibilities on it. And practically I have no ability to enforce the terms of any licence once the code is in public anyway, so why should I spend time engaging in the licencing theatre
"Slap a GPL3 in there" is pretty much the opposite of what the parent commenter wants, though. At 5500 words, GPL3 is a long read - I have a rough idea what it says, but can't say I've ever read and understood every word. Making my users do so might constitute cruel and unusual punishment :)
0BSD looks short and sweet, though. I don't need people using my libraries to include my LICENSE.txt, or give attribution, or tag any changes with their own copyright notice. Just take the code and do whatever you want, no strings attached.
> Making my users do so might constitute cruel and unusual punishment
The very opposite. GPLv3 is written by lawyers and for very good reasons. The more detailed a license is the more clear it is. That's why it clearly defines what constitutes distribution and so on.
This is also why things like rent contracts have long lists of what is allowed and expected and what is not, and when and why and so on: it creates clarity.
Clarity means less debating in court.
If you want legal trouble, sign a very short contract.
> If you want legal trouble, sign a very short contract.
If the length of contract is what is protecting you, you're already in court. If it's a jury trial, all bets are off no matter what the law or contract says.
> If the length of contract is what is protecting you, you're already in court.
Not at all! A strong, detailed, clearly written contract reduces the amount of "wiggle room" for litigation.
The other party is much more likely *not* to sue you in the first place if they know they would be fighting an uphill battle against a strong contract.
That's why GPLv3 has anti-patent-troll clauses for example.
The problem with most licenses is that they require that the license be distributed with the content that is licensed. I don't want to force people to do anything. I want it to be as if copyright doesn't exist.
By default copyright exists so by doing nothing you are copyrighting your work and preventing people who want to be in the legal clear from being able to use it (OSS games for example).
The best you can do is watermark your art with CC0 and/or include the license or a link to it in the file metadata.
But pretending we are living in a post-copyright utopia severely limits the reach of your work.
I'm not pretending copyright doesn't exist, but I want other people to be able to interact with my work as if copyright and other forms of IP didn't exist.
There doesn't seem to be a perfect way to achieve this. The problem with CC0 (and I believe the reason OSI doesn't approve of it) is that it says:
> No trademark or patent rights held by Affirmer are waived, abandoned, surrendered, licensed or otherwise affected by this document.
Whereas many licenses can be interpreted as an implied patent license[0]. I guess adding an additional declaration saying something like this could fix that:
> All trademark and patent rights associated with this work are permanently waived.
But I have no idea if that would work and it's frustrating that there is no simple way of opting out of intellectual 'property' ownership all together.
The closest, at least when it comes to software may be 0BSD considering large corporations like Google are willing to accept it.
The right copyright gives is a monopoly on distribution and on granting of conditional permission to distribute. The right is the monopoly itself, not the ability to do it. If you waive that right, you're saying that anyone is allowed to distribute and to grant permission to distribute. An author of a work in the public domain can license their work to someone under any license -- as can anyone else.
Freedom to me is whatever exists in the absence of coercion. GPL licenses depends on coercion, so they cannot produce freedom as I see it.
To illustrate this, imagine an alternative universe where violence is impossible. In this universe IP as we know it including copyleft licenses couldn't exist as violence or credible threats of violence by a state (or similar entity) is required for enforcement. I want to get as close to this hypothetical universe as possible by avoiding everything that depends on violence and coercion as far as possible.
The only thing that the GPL family prevents is coercion, effectively reaching a local minima of coercion-by-others as a function of coercion-by-vendor.
Granting the freedom to coerce doesn't reduce coercion.
Not distributing source code is not coercion. GPL forces people to do work in exchange for a promise that they will not be subject to volience, that's coercion.
You're right that in a universe where everyone grants their end users necessary freedoms and don't use coercion to limit those rights, GPL would be completely unnecessary. Until that happens, coercing people to stop further coercion is the best thing we have.
I'm taking about a universe where violence is impossible. Not a universe where people are compelled to distribute source code whenever they have distributed a binary file.
I'm fundamentally opposed to the idea of forcing people to do what I want and that's the clear purpose of GPL.
Somebody not doing what I want is not coercing me. I'm free to use or not use a binary distributed without it's source code, or I can try to reverse engineer it. Forcing them to do the additional work of sending me the source code would be coercing them, it would be impossible in my hypothetical alternative universe, and it would be wrong in this universe.
Your definition of freedom seem to be other people doing what you want them to do. That's not how I see freedom. Someone offering software as a service is not a threat to anyone's freedom. Preventing them from doing so by the threat of state violence makes both the business and it's potential customers less free.
I've used CC0 when I want to free downstream users to reuse the code in any context and under any license of their choosing, free from the burden of attribution or preservation of my copyright notice. It's great for code samples.
I don't. I don't want any license at all, I just want people to be able to use what I have written in a natural and completely bullshit-free legally unencumbered manner. Unfortunately, copyright law has forced me to declare this outright in what is, apparently, convoluted and tedious legalese.
Because software licenses are a mess Linux and Samba implement the same protocol UT can't share code as one is under GPLv3 and one is under GPLv2. In the past GPLv2 was the universal receptor license for open source licenses (is that license GPLv2 compatible and my code is then I can release the result as GPLv2). GPLv3 changed that. Now there is only the universal donor. There is turns The public domain. However that one has been utterly FUDed by boomer lawyers being compared to littering by the highway. So now we need a rubber stampable (so people can use it until copyright gets killed ... Uhhm ... fixed) public domain equivalent license. OBSD does that without "any uncertainty" and it being a "proper license".
It's not "boomer FUD". Germany and Japan currently won't listen to you if you say your work is public domain. They only respect copyright expiration and licensing agreements. So if you say your code is PD, and someone in those countries use it, you can still sue them for copyright infringement and possibly win. If you retain copyright and use literally any license then the people using your code have a much stronger defense against you.
You might not care about Germany or Japan, but there's plenty of FOSS developers there that need to have working licenses. Is it possible that those jurisdictions might change their mind? Yes, but it's also unlikely. Copyright harmonization has been entirely focused on the needs of the proprietary world[0]. Thus, if your goal is to effectively disclaim copyright, you need to choose legal actions that fit within the mold that the proprietary world has forced upon us. If the proprietary world does not care about artists being allowed to dedicate works to the public domain, then you won't be able to do so internationally.
[0] e.g.
- Very long copyright terms
- Unwaivable moral rights for (copyright-wielding) artists
- Strong domestic enforcement for international copyright holders
- Upper bounds on exceptions to copyright
- Forcing everyone to adopt DMCA 1201
- Making copyright exempt from normal democratic process
I've been frustrated that there exists no lawyer-verified, internationally-viable way to permanently waive all claims and rights to things (software or otherwise) I publish, and to assert that those rights will not be later reclaimed (whether by me, parties representing me, or heirs or successors who might claim to have inherited my rights).
CC0's biggest issues for me are that, one, it explicitly does not waive rights to patentable ideas contained in the work:
> No trademark or patent rights held by Affirmer are waived, abandoned, surrendered, licensed or otherwise affected by this document.
...and two, that it only specifically says that I will not revoke the license or undertake "any other legal or equitable action to disrupt the quiet enjoyment of the Work by the public". I don't just want to enable "quiet enjoyment", I want to permit every legally permissible use of my work (anything I could grant a license for). "Vocal profiteering" is arguably not "quiet enjoyment", but I want to ensure that neither me, my representatives, nor my heirs or successors, could make that argument to revoke the license I grant.
The Blue Oak license nicely disburses patent rights:
> Each contributor licenses you to do everything with this software that would otherwise infringe any patent claims they can license or become able to license.
...but expressly retains copyright (preventing someone else from redistributing the work under different terms), and requires that the work always be distributed with a notice that the work is licensed https://blueoakcouncil.org/license/1.0.0
I wish to completely unburden anyone who finds my work desirable from the burden of attribution. This is why Public Domain is the end goal, and why licenses that do not expressly attempt to un-copyright my work are insufficient. MIT0 merely grants permission "to deal in the Software [to say nothing about non-software work, or aspects of a work] without restriction", and 0BSD is even weaker, granting only "permission to use, copy, modify, and/or distribute this software for any purpose with or without fee" [nevermind any activities that aren't expressly listed in the license, but which might be restricted by a country's system of copy-, artists', or moral rights, or any other restrictions I may impose besides fees].
"Quiet enjoyment" is a archaic legal term of art, arising (I believe) in law governing conveyances of real property. The Black's law dictionary on my desk defines "quiet" in this context as "Unmolested; tranquil; free from interference or disturbance." That is, it's not a condition imposed on the transferee ("if you do not enjoy the conveyed property quietly, I can make trouble for you").
The term causes enough confusion to be retired, but lawyers do love their stock phrases.
> way to permanently waive all claims and rights to things (software or otherwise) I publish, and to assert that those rights will not be later reclaimed
Probably because to do the second, you can't do the first - you have to retain enough rights to keep someone else from copying and relicensing it. And if you do the first, you can't do the second.
The best way probably involves a third party, like a foundation or trustee, taking on the ownership under a contract that they will never switch ownership until it's naturally released by time. Of course, that costs money to maintain, otherwise it becomes an asset that gets sold off.
Relicensing from second parties to third parties is acceptable, because so long as my publication is discoverable (via catalog, search engine, etc...), the third party could obtain the public-domain permission for (the copyrighted parts which are in fact) my work.
Somewhat analogously, in the United States you can publish copyrighted works that contain, and even largely consist of, public-domain content. An example off the top of my head is this book [ CliffsComplete Shakespeare's Hamlet ]( https://www.amazon.com/CliffsComplete-Shakespeares-Hamlet-Wi... ) that I used in school. This was how I was introduced to, and first read, the play. The bulk of the text consists of Shakespeare's "Hamlet", which is in the public domain in the United States, and does not require attribution or licensing. The annotations, footnotes, and explanations are original and copyrighted by the publisher (Cliff's Notes). However, I can extract the original play within and redistribute it, or novel works based on it, without worrying about violating rights held by Cliff's Notes.
When anyone is allowed to redistribute under their own terms and attribution, what the user can or can't do depends entirely on where/who they got the work from. Which means in your case that if the place you published the work becomes inaccessible, for example the heirs or successors stop paying for hosting, people that got it from you using the hypothetical super permissive license might no longer be able to prove they are allowed to do what they want with the work.
That is "merely" a matter of record / archiving. I'd be happy to annotate my first-party copies with copies of the license, so there would exist copies of the work with the license / public-domain declaration attached. The authenticity could be attested by archived copies (perhaps held by Github, Wikipedia, the Internet Archive, or web scrapers / archivers like the WayBack Machine, Google Archive, etc.), or by being signed with a cryptographic key of mine.
This same problem exists with the public domain. If, in the United States, you find an unpublished century-old manuscript in your attic, its copyright has expired, and you can't claim rights to it. However, because nobody else knows about it, you could in practice type it all in to a computer and claim to have authored the work.
edit: This record also does not have to be perfect. I will probably be dead in a century and unable to care about, or further influence, the usage of my work. Under current U.S. law, my works will already be in, or eventually entering into, the public domain. If you really want to be long-term, there's no guarantee that in the distant future anybody will have a copy of, or care about, my works. Alternatively, there is no guarantee that, in the long-term future, the present state of intellectual property law, or any actions I take within it, will have any bearing on the rights future beings will have to do or not do things with my works.
Personally, I think the fact that we even have to talk about the suitability of Unlicense vs CC0 is evidence of how terribly stupid copyright law is. I wrote code, it isn't fancy or interesting and I make no money from it nor do I intend to, and I just want to let everyone know that I truly do not care at all what they do with it. Why is it so hard to let people know that they can use information in the natural way that information is used?
If you're in the US, you can release works into public domain by writing "This is public domain".
In other countries, public domain doesn't exist and in order for someone to use a work, there needs to be legal proof that you have the right to use it. Hence the requirement for licenses.
In a lot of ways, I prefer the latter approach because it puts the burden of proof on the copier, not the creator.
Nit: The public domain simply means a body of works that are outside of the copyright protection and thus exists for every country with the expiration of copyright, though it might not be named so. What those other countries actually lack is either the dedication to the public domain before the expiration or the inability to waive moral rights.
I don't think this is a problem outside of corporate scenarios. If you just explain your position in plain human language, like you just did, then reasonable humans without a legal department will be able to comprehend it and act accordingly.
I use it and will continue to use it because it most accurately reflects my intentions for my software. If some organization can't comply with it, it's worse for them. They should change, not me.
OK, they should change. Great. They won't, though, so your code won't be usable by them. If that's not a problem, then your code was never really relevant to this discussion. If that is a problem, then shoulds and woulds ain't gonna help.
The purpose of freely distributing code is not to be useful to corporations. It's to foster a spirit of sharing with the community.
If I put something up for free, it's because I want other programmers to be able to freely use it. I don't care if a big corporation with expensive lawyers likes the way the license is worded.
But if your license is stupid, then other programmers can not use it. Congratulations.
There is really no excuse for not at some point in your entire programming life, taking the time to go over these things, figure out what they end up meaning in real life, and thereafter knowing which ones actually do represent your wishes and intents. Picking one whos only property is it's so short it doesn't actually do anything hardly counts.
It doesn't really matter how big and complicated a license is any more than how big and complicated a compiled executable is or all the parts in a car. What matters is does it do the job that needs doing. All else being equal, smaller and more elegant is better as a direction or principle of course, but the reason we even have writing is to assemble work into a reusable package, so that time-consuming work like writing a book only has to be done once, and then everyone else gets to use the big complicated work many times over without having to re-create it each time.
It's an efficiency and a power amplifier to be able to pack up a bunch of complicated things into a writing and then treat the writing as a single simple thing.
You happly use gcc (or your car, or whatever) many times every day. You did the work one time to figure out that the tool you need is gcc, and after that, all you mentally think about is just "gcc" not the thousands of lines of code or the millions of machine instructions that make it up every time you use it. If I write a new "The Un-cc" that only has about 8 lines of code and is oh so refreshingly simple to understand, one would hope that you would not use it.
The point of the established and thorough licenses is exactly to do a whole lot of hard work once and let everyone else reuse it countless times.
The point of a stupid new license that pointedly and intentionally does not do any of that hard work is there is no point at all to it. It's just a stupid idea, and as such, it's probably no great loss that other people can not use code from an author who chose such a stupid license.
It's not about "some organizations being unable to comply", it's that it is poorly written and thus incompatible with the legal systems of countries. But screw everyone outside of the US I guess
In the License section I write this text - "Public domain, see the LICENCE file.", and in the text of the license itself the word "unlicense" is also nowhere used.
Then I would not be able to use your code no matter my intentions no matter how well meaning you are.
In my country you cannot dedicate something to the public domain (it happens automatically 70 years after your death).
The Unlicense would most likely allow you or your heirs to come after me and sue me for copyright infringement for many decades.
If you (pretty much) only want US citizens (or naive people) to use your code that's fine. However given that you choose this license it seems like this wasn't your intention..
"You can't dedicate something to the public domain in my country" - that's the whole point of Unlicense. Yes, I can do it, in my country, in your country, in any other country, and even on Mars. And I don't need an approval of any authority to be able to do it. The problem is not that I can't do it, the problem is that countries refuse to acknowledge that fact.
How does the Unlicense most accurately reflects your intentions? The Unlicense has three different components: grants (paragraph 2), no warranty clause (paragraph 4) and the dedication to the public domain (paragraphs 1 and 3). Everything but the PD dedication is common to other PD-equivalent licenses so I believe you have two intentions:
1. You want the PD dedication whenever it works.
The dedication clause is unfortunately most problematic in that, for example, it never works in jurisdictions where copyright laws are recognized but no actual dedication to the public domain is possible. The complexity of CC0 solely exists to make it effectively PD-equivalent even in such cases.
2. You don't like "lawyer speaks" and prefer shorter licenses.
Okay, unfortunately CC0 is bulky and while legally absurd I can somehow relate to that line of thought. But does that mean the PD dedication clause should exist in the license itself? No! You can easily make a PD-like license by writing your own dedication plus very permissive and short license like zero-clause BSD. In this way your intention to the dedication remains explicit (or even stronger) and you can pick legally safer licenses. Indeed this is my preferred method for the dedication [1].
Also remember, the actual SQLite "license" [2] from which the Unlicense claims to be inspired is not the license. It is just a dedication and words of blessing. The actual license, in case the dedication doesn't work, is available for purchase elsewhere [3]. The Unlicense authors are seemingly ignorant of this fact.
That old gcc is too big and complicated. In fact all the normal c compilers are too complicated. No one can read them and understand everything. There's thousands of lines of complicated code in there. Even in new reimplimentations like llvm.
So I have a new cc that only has about 8 lines of code. It's refreshinly simple to understand.
Hum... Rational (from the point of view of the entire enterprise) bureaucracy bodies probably exist, but they are not common. But that second part "something that could happen to anyone" is completely unfounded. It may be because of something that could happen to you, or it may not. There is no reason to assume either way.
As people keep ignoring on the context of software architecture, you are not Google.
Everyone has different reason to publish their code using open-source license. I like sharing my code but I don't like when multi-billion dollar company make a profit from it for example. Ofc, that's what I agreed to when I release my code under open-source license, but we are all free to have different sentiment about it.
A great talk on software licensing made by the creator of the 0BSD license (Rob Landley) and the history behind the license is here https://youtu.be/MkJkyMuBm3g
Good call. I had somehow quite forgotten about it, though I know I read all the N-clause BSD variants a decade or so ago. I’ve added a section at the end with that as an alternative recommendation.
Google uses Sqlite, which is simply public domain. Did someone at Google actually purchase one of their pseudo-license 'warranty of title' documents they offer for companies (supposedly mostly in other countries that officially don't recognize PD grants) with more legal budget than sense? Google also uses SBCL, which is mostly public domain "in jurisdictions where this is possible, or under the FreeBSD licence where not." Does the entry for it in the list of third party software say Google is using it under the FreeBSD license or PD? (They also employ at least one developer who is a core maintainer on the side.)
The legal department's default answer is "no" to anything they haven't already said "yes" to, so it wouldn't surprise me if your default options are limited. But presumably with useful enough software under other PD-equivalent things they could be forced to say "yes". As an aside I'd be very interested to know the extent of internal audits Google does on its code for legal risks in un-cleared-by-legal dependencies, dependencies-of-dependencies, and presence of code derived from stack overflow/popular books/random places on the internet indexed by google.
A bit off topic, but commenting on footnotes is fair play: I don't believe that 'let' is in fact autoantonymic.
A let is more-or-less lost, as a noun in English, but in the form inlet and outlet it remains ubiquitous. In "let or hindrance" we might observe that a right of travel (a let) might well be a hindrance to other uses of the property.
A let in tennis is so-called because you 'let' the server try again. The sense is clearly one of permission to the player, not hindrance by the net.
Your objection interests me, because these are both cases that I figured out myself with much contemplation and was quite content with many years ago and have never subsequently looked up in any reference (and I doubt your alternatives ever occurred to me). But after a spot of investigation now, I don’t believe you’re right on either count and stick to what I have written, at least for now.
https://duckduckgo.com/?q=let+tennis+hinder: the consensus among the first few results, which are reputable-looking sources, seems to be that the hinder sense is probably correct, though some suggest as an alternative the French filet meaning net.
In my response I was addressing the specific phrase “let or hindrance”; that let has historically been auto-antonymic is incontrovertible and well-documented and (whether rightly or wrongly) I was interpreting samatman’s comment as referring to the two specific contexts where I say it is still exhibits its historical auto-antonymicity.
I was. I'm still convinced that the folky etymology we've found online for tennis is wrong, because "serve, let, fault" are all about the server. But that's hardly relevant.
If you want an approximately-public-domain license that pokes fun at the officious and arbitrary ideas lawyers hold, and that those lawyers say isn't officious enough, you want WTFPL.
If you want an approximately-public-domain license that seems pretty sincere but the same lawyers say isn't officious enough, you want Unlicense.
If you want an approximately-public-domain license that lawyers seem to think is quite nice, but that's more for cultural work than software, and whose creators don't recommend it be used for software, you want CC0.
If you want an approximately-public-domain license with a much more famous sibling, and plenty of lawyers say the sibling is good, you want 0BSD.
Badly. It is absolutely unclear how a judge would interpret it when the licensor later says he intended something else then what you interpret to be the spirit of the license as there is no "word of the license" that is legally parseable.
The way people and organizations stand in the way of public domain releases is beyond rude to the point of genuine evil. My code is just code. It should be possible to let all others read it without obligation. There must be some way to restrain this relentless desire to cocoon everything in layers of legal entanglement. It is just code and I as the author desire others to be able to read it. This is not a crazy plot or rocket science.
There might be technical problems with the name itself, but the license text explains the intent and refers to the https://unlicense.org/ site which has additional clarification. Proper use as recommended on the site has zero legal ambiguity.
The idea that "BSD0" or "CC0" are better names makes no sense whatsoever. Public domain releases are made by those who do not want the baggage of a BSD or Creative Commons license. Maybe someone in these organizations should talk to someone familiar with marketing and namespaces to come up with a real solution?
The CC0 suggestion is especially ironic since that has not approved. It doesn't actually exist as a legal standard which makes it unappealing.
The OSI, FSF, FANG and other such companies, and the European Union need to get over the idea that they can and even must dictate all aspects of intellectual property ownership as they see fit. My code is not theirs and this is the core issue. My code is mine and thus mine to give away completely free without being encumbered by licensing schemes intended to yield control or profit. That their schemes for control and profit are so out of control that there is no agreed upon or even consistently imagined escape is a sign that they are operating beyond rational limits.
> My code is not theirs and this is the core issue.
This is _exactly_ why a lot of jurisdictions assign/assume copyright by default, and make it hard to 'remove'. Otherwise authors would see their work ripped off in a lot of inventive ways.
When the default state is that every work is copyrighted, that removes a lot of possible confusion, and possible excuses.
Right. Under the older copyright regime in the US, you more or less had to add the appropriate incantation/symbols and send a copy to the Library of Congress or you pretty much had no leg to stand on collecting compensation if Disney or whoever decided to rip you off. The US decided that wasn't a good thing and that works should be copyrighted automatically. (Registration still has some benefits for collecting on copyright claims.) But that also means you now need to get rid of that automatic copyright, assuming you can, if you want to do so.
> There must be some way to restrain this relentless desire to cocoon everything in layers of legal entanglement.
It's very simple. We, in most countries, live in a legal system where creations default to maximum IP protection rather than minimum IP protection.
Everyone has their own opinion about that - and it's fascinating to read all of them - but this is the world we live in. Some people seem to have trouble understanding that and have retreated to the imaginary land of Shouldistan.
> Proper use as recommended on the site has zero legal ambiguity.
See above: we are not in Shouldistan.
> The CC0 suggestion is especially ironic since that has not approved. It doesn't actually exist as a legal standard which makes it unappealing.
They are words. They have meaning. The licence text doesn't need to 'exist' in whatever ontology you've dreamed up. You can write your very own licence and it will have legal effect by dint of the fact that words have meaning.
> The OSI, FSF, FANG and other such companies, and the European Union need to get over the idea that they can and even must dictate all aspects of intellectual property ownership as they see fit.
The European Union can dictate all aspects of intellectual property ownership as it sees fit[0]. The others don't, nor do they think they can, because that would be stupid, and none of those groups are (that) stupid. I don't know what you're referring to in suggesting that they believe that. I'm not altogether sure that you do either.
Again, you (very likely) do not personally dictate the law of the land where you live, so you'll have to suck it up and write a licence that conveys the legal restrictions (or lack thereof) that you intend to apply to your code.
I'm sorry that we don't live in a land where the law is exactly as you want it to be. You can stay in Shouldistan, raging against the fact that the world isn't exactly as you would like it to be, or you can get smart, grit your teeth, accept that the world is imperfect, and negotiate it as best you can to achieve your goals.
[0] ...more or less, in most member states, for the purposes of this conversation.
Why does the concept of "public domain" actually matter though?
It's clearly a legally frought concept, and nobody actually believes in public domain itself as an ideal (as far as I know), its just seen as a means to an end, the end being a release of all "legal entanglement" upon the work.
If that means is problematic, why not use a less problematic means?
Specifically, there exist licenses that give everyone every possible right that they can be given without needing to invoke public domain at all, and that should be sufficient for anyone who doesn't want to retain their default rights, so why is public domain anything more than a legal dead-end? (ignoring of course that it's still relevant as the eventual end of copyright once someone's been dead for longer than Walt Disney)
I am largely ignorant to license law so forgive this probably basic question
If the idea of public domain is clearly established in law should it not be enough to have a license file that simply reads.
Everything in this repository is public domain, all contributions to this repository are public domain.
Obviously if someone were to submit code stolen from a project under another license it would create an issue. But the same could be said for any project.
The concept of 'public domain' in the sense of all legal property rights and responsibilities being relinquished is not clearly established everywhere, and where it is, it sometimes applies only in certain circumstances (like years after the death of the creator).
I'm not contributing to open source projects privately right now. But if I was publishing code on my own, I would probably make it actually unlicensed at first to keep my options open as much as possible. I would just put it somewhere and say: "here is code, you are technically able to look at it, but I am not giving you any permission". If somebody wants to use it and needs to follow copyright (for example in a commercial context) then I can grant them a free or paid license.
I like the aspect that people who don't care about copyright have an advantage, and people who have internalized it cannot even use the code privately or look at it. I think it tells that I am not a big fan of the idea of intellectual property...
> I like the aspect that people who don't care about copyright have an advantage, and people who have internalized it cannot even use the code privately or look at it. I think it tells that I am not a big fan of the idea of intellectual property...
I can’t stop you doing that. I’m not sure I even want to. I hate copyright, and while I’m not sure it should be completely destroyed, certainly it should be weakened by — say — ninety percent.
However, that doesn’t mean I’m willing to break it. What I’ve internalised, having seen the alternative, isn’t copyright — it’s the rule of law. I’m not willing to break that just because it’s personally inconvenient.
You are probably not very old but anyway . Let's say you die in accident. Nobody will be able to use your code until either copyright is repealed (the boomers half life has kicked in long enough such that they are irrelevant and we can change copyright) or you are 70 years dead.
I understand you wanting to keep your options as long as possible. I see three options for you:
- leave it without license and hope you will be able to license it as you want. Take no contributions. Have total freedom. Never accept contributors.
- pick any non public domain equivalent license and require all contributions to allow you to relicense their contributions in the future. When you change the license in the future the all previous version released under that license stay that way.
- pick a public domain equivalent license 0BSD preferably. All contributions under that license means you and their code and be relicensed by anybody including you. If you later decide to relicense you can do so without needing the contributors ok. Old versions stay in the public domain. Newer code (after relicensing doesn't).
That'd mean:
* OSI approves it
* FSF approves it
* Google and other companies with large stakes in FOSS find it acceptable
* Github + Gitlab support it in their templates
Can we have that?