Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Unlicense is an inferior license wrapped in an atrocious name

chrismorgan.info

Post reply on HN
Print view View original

131–140 of 185 posts

Re: The Unlicense is an inferior license wrapped in an atrocious name

#131

Post by goodpoint »

aloe_falsa wrote: post #82
smoldesu wrote: post #47

Earlier quoted context omitted.

// Oops, I misunderstood the context of this discussion. Disregard this :p

"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 t…

> 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.

Re: The Unlicense is an inferior license wrapped in an atrocious name

#132

Post by avhon1 »

falcolas wrote: post #128
avhon1 wrote: post #123

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 righ…

> 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 tr…

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.

Re: The Unlicense is an inferior license wrapped in an atrocious name

#133

Post by jobigoud »

avhon1 wrote: post #123

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 righ…

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.

Re: The Unlicense is an inferior license wrapped in an atrocious name

#134

Post by goodpoint »

Pet_Ant wrote: post #15
683686285 wrote: post #13

Earlier quoted context omitted.

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.

> pretending we are living in a post-copyright utopia severely limits the reach of your work

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.

Re: The Unlicense is an inferior license wrapped in an atrocious name

#136

Post by avhon1 »

jobigoud wrote: post #133
avhon1 wrote: post #123

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 righ…

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…

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.

Re: The Unlicense is an inferior license wrapped in an atrocious name

#137

Post by mattarm »

kemitchell wrote: post #120
hannob wrote: post #19

I wish people would get together to agree upon a public domain alike software license that everyone finds acceptable. 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?

Blue Oak Council, a gaggle of savvy open licensing lawyers, published a model permissive software license: https://blueoakcouncil.org/license/1.0.0 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 directo…

> 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.

Re: The Unlicense is an inferior license wrapped in an atrocious name

#138

Post by kemitchell »

mattarm wrote: post #137
kemitchell wrote: post #120

Earlier quoted context omitted.

Blue Oak Council, a gaggle of savvy open licensing lawyers, published a model permissive software license: https://blueoakcouncil.org/license/1.0.0 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 directo…

> 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 lic…

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....

Re: The Unlicense is an inferior license wrapped in an atrocious name

#139

Post by BeFlatXIII »

goodpoint wrote: post #131
aloe_falsa wrote: post #82

Earlier quoted context omitted.

"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 t…

> 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 clar…

> 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.

Re: The Unlicense is an inferior license wrapped in an atrocious name

#140

Post by jobigoud »

kemitchell wrote: post #120
hannob wrote: post #19

I wish people would get together to agree upon a public domain alike software license that everyone finds acceptable. 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?

Blue Oak Council, a gaggle of savvy open licensing lawyers, published a model permissive software license: https://blueoakcouncil.org/license/1.0.0 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 directo…

From the license text:

> 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.

Post reply on HN