[License-review] New License for Consideration - Public Benefit Zero Copyright License v. 2.0

[Rob Landley]

On 12/18/24 05:46, Carlo Piana wrote:
> Kevin, you are right. This license is both dedication to public domain AND a copyleft license (see clause 8 in the Definitions sections or 1, 2 in the license grant part).
> 
> I think this does not pass the minimum requirements for being considered at all. For starter, it admittedly has not been reviewed by a lawyer, and I doubt that a lawyer would have permitted:

I also note that at least in theory, "public domain equivalent licenses" 
should be fungible (sort of the point) and there are already a 
half-dozen common ones listed on the wikipedia page for the concept:

https://en.wikipedia.org/wiki/Public-domain-equivalent_license

I did Zero clause BSD, which is used by Google and Microsoft and 
Wikimedia, and about 60k projects on github:

https://github.com/search?q=license%3A0bsd&type=Repositories&ref=advsearch&l=&l=

0BSD was my explicit attempt to make a public domain equivalent license 
that appealed to both corporate and hobbyist interests, while being 
small and simple enough for laymen, and it seems to have worked:

https://landley.net/toybox/license.html

I had to manually walk it through SPDX and Github's choose-a-license 
approval processes, and the main reason I'm still subscribed to this 
list is because I had to convince OSI (which had retroactively renamed 
0BSD in a way that caused issues) to put it back to the original name 
_twice_ so far.

This area gets a lot of bikeshedding ("your copyright license only 
addresses copyrights, not patent and trademark in an internationally 
compatible way in under a hundred words", "Kirk McKusick personally 
giving you permission to call the OpenBSD suggested template license a 
BSD license isn't good enough"), but fundamentally people still use code 
with dedications like:

https://github.com/mkj/dropbear/blob/6ae3a09ef33b/libtomcrypt/LICENSE
https://github.com/mkj/dropbear/blob/6ae3a09ef33b/libtommath/LICENSE

And then stick their own licenses on them years later for comfort:

https://github.com/mkj/dropbear/blob/master/libtomcrypt/LICENSE
https://github.com/mkj/dropbear/blob/master/libtommath/LICENSE

Without explicitly admitting "I didn't get MORE permission from the 
original copyright holders, so I'm still technically depending on the 
earlier permission grant, I just don't want it to LOOK like I'm doing so".

Yes the second libtommath link above is a new hand-rolled page long 
license not listed on the wikipedia page above, for the one project. 
There's a ton of these. because:

https://xkcd.com/927/

Most people don't really seem worried that a public domain declaration 
isn't good enough, they're worried OTHER PEOPLE won't think it's good 
enough. For example https://en.wikipedia.org/wiki/Qmail didn't have ANY 
license statement from 1998 through 2007 but was still one of the most 
popular mail transfer agents. Similarly, Arch Linux just switched its 
package sources to 0BSD last month after a couple decades of "no license 
specified":

https://linuxiac.com/arch-linux-adopts-0bsd-license-for-package-sources/

Anyway, this bit's one of my stomping grounds, thought I'd say hi.

Rob

P.S. During the github choose-a-license approval process I laid out the 
rationale for 0BSD and the issues I was trying to address at some 
length, if you're curious:

https://landley.net/notes-2017.html#26-03-2017
https://landley.net/notes-2017.html#27-03-2017