The Unlicense is an inferior license wrapped in an atrocious name
chrismorgan.info
chrismorgan.info
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?
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.
Don't tell Rob Landley that (the creator of the license). He is quite adamant that he used a BSD license as the basis.
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.
Github is the de-facto default sourcecode plattform. If it isn't a standard option on Github then it doesn't feel like a standard license.
There are really a lot of times that this is not important, or may be detrimental to your projects success.
There's a brief write-up here: https://blueoakcouncil.org/2019/03/06/model
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.
I'm executive director of Blue Oak Council.
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 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....
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.
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.
"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.
A one-line sentence should be more than enough to clarify my intent.
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.
> 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.
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.
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:
https://wiki.creativecommons.org/wiki/CC0_FAQ#May_I_apply_CC...
Something like this:
// Available under http://creativecommons.org/publicdomain/zero/1.0
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.
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.
What's wrong with 0BSD?
https://opensource.org/licenses/0BSD
"Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted."
Not being a lawyer I can only speculate on other problems. It doesn't get around the problem of moral rights, so probably doesn't work in France.
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".
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.
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).
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.
(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.
2) Actually, SQLite has this: https://www.sqlite.org/purchase/license
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.)
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.
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.
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.
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 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.
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 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.
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.
And severely limits the freedom of end users!
GPLv3 protects around tivoization and proprietarization. Other licenses do not.
The net effect is that end users are unable to benefit from the openness.
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.
Granting the freedom to coerce doesn't reduce coercion.
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.
These are all things that currently threaten the freedom of users and developers.
Ramblings about impossible alternative universes adds very little.
> End result is the same without the uncertainty.
You fell for the boomer FUD.
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
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].
The term causes enough confusion to be retired, but lawyers do love their stock phrases.
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.
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.
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.
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.
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.
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.
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.
Imagine two people from a corporation having this conversation:
'Can we use this?'
'Yes, it is unlicensed'
'So we cannot, we don't license it'
'No, we can'
'But you said...'
and so on...
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..
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.
[1] See https://github.com/lifthrasiir/rust-strconv/blob/master/LICE... for the example. (It eventually got into the Rust standard library, hence weird triple licensing.)
[2] https://www.sqlite.org/cgi/src/file?name=LICENSE.md&ci=trunk
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.
This makes it better. Enjoy!
As people keep ignoring on the context of software architecture, you are not Google.
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.
<LICENSE>+GOAWAY
This license must never be enforced.
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.
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.
https://duckduckgo.com/?q=etymology+of+let+and+hindrance: not much obviously canonical, but https://www.phrases.org.uk/meanings/without-let-or-hindrance... at least sounds reasonable, and agrees with me, including a citation in the last paragraph from 1799 which is unequivocally being used in the sense I nominated.
It's an interesting phenomenon, socially. But legal-technically, it's far off the state of the art.
https://github.com/pygy/The-Romantic-WTF-Public-License
(I know it's a baaaad idea, I still love it)
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.
Gag...
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.
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.
> 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.
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)
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.
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.
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).