[License-discuss] Request for Comment: Software and Development License, version 3.0

Tue Jan 30 00:58:20 UTC 2024

On Jan 29 2024, at 5:47 am, Kevin P. Fleming <lists.osi-license-discuss at kevin.km6g.us> wrote:
> On Sun, Jan 28, 2024, at 14:47, Alec Bloss wrote:
>
> > Hi all,
> >
> >
> > I wanted to ask for comments/input on a license that I've been working on, prior to considering sending it for an official review. It's aim is to be as widely compatible with other open-source licenses without some of the compatibility issues of the GPL licenses or extra requirements of Apache 2.0 license (notification of modified files required), etc. The full text of the license in its current draft is below.
> >
> > Thanks in advance for your input and all the work that you do!
> >
> > Alec
> >
> >
> >
> > Software and Development License version 3.0 (SADLv3)
>
> This is a nit, but this name is problematic in my opinion. You can't license 'development', as that's an activity.

I would tend to agree that you are correct so I'll most likely alter the name. I would like to point out for reference that the CDDL is a OSI-approved license and uses Development in its name (doesn't mean that's a good thing however).

On Jan 29 2024, at 11:06 am, Nicholas Matthew Neft Weinstock via License-discuss <license-discuss at lists.opensource.org> wrote:
>
> Can you provide a link to SADL v 1.0 or 2.0 for reference?
>
>
> -Nick

I apologize, I should have clarified this in my original email. Both prior versions have problems I don't like (for example, v1 has a patent-retaliation clause that is excessive for what I wanted to achieve, v2 #5 is too broad, v6 has examples that are not necessary or open-ended etc.), and I do not intend to submit them for review, but for context they are linked below.
https://centre.libranext.com/libranext/llp/-/blob/main/SADLv1.md - SADL version 1
https://centre.libranext.com/libranext/llp/-/blob/main/SADLv2.md - SADL version 2

On Jan 29 2024, at 1:13 pm, Josh Berkus <josh at berkus.org> wrote:
> On 1/28/24 11:47, Alec Bloss wrote:
> > The Software and Development License was created to fill what
> > was perceived as a void in open-source licenses. The GPL licenses, while
> > great, can and have caused unnecessary incompatibilities with other open-
> > source licenses. On the other end, the BSD licenses, while widely
> > compatible, are very permissive and do not have some of the protections
> > provided in other licenses. The SADL aims to be widely compatible while
> > still maintaining some protections.
>
> This "middle road" was the goal of both the Apache and Mozilla licenses.
> While you've taken a different tack on it, it might be worth thinking
> about how APL/MPL approached it and why.
> More comments:
>
> 1. There's no patent grant in this license.
> 2. The AI/ML permission clause is interesting, but likely unenforceable.
> The courts will let us know in the future.
>
> 3. The provision for using "3.0 or later" of the SADL necessarily
> requires that their be an authority to determine what is an official
> version of the SADL, which will need to be part of the license.
>
> 4. The linking language is clearly intended to prevent this license from
> being incompatible with other open source licenses (IANAL, so I can't
> say if it works). However, that's not bidirectional; other OSS licenses
> will still be incompatible with this one due to *their* terms. Also,
> F.2. would be incompatible with several licenses.
>
> 5. That linking language would work with proprietary and
> domain-restricted licenses as well, as long as those permit
> redistribution of source. While that's not necessarily a blocker to
> approval, you might want to think through the implications.
>
> 6. It seems weird to call the first release of a new license "3.0".
> --
> Josh Berkus
>

I have issues with Apache license, but the MPL is close.
There is a patent grant in B.3

I agree, this clause my be unenforceable depending on how courts rule. I believe it is harmless in either case. If it is not enforceable it will at least be a reminder to anyone reading the license that it's good practice to give credit in many cases, and hopefully that will lead to more citations in AI works even if not required.

Good point, I will address this in the next revision.

It is intended to provide as broad compatibility as possible, but I don't believe it makes it universally bidirectional (unless this is at some time in the future proven by court). However, it allows as many cases as possible, intentionally, or at least creates as few infringing situations as possible with the SADL licensed software. If you have an example of a license that isn't compatible with F.2 I would like to know so I can consider those cases.

I understand my views on mixing proprietary/open source software isn't shared by everyone. There is software I write that I would like to be usable as widely as possible (and some that I prefer to be strictly open-source, but in those cases I use a different license). This is slightly more restrictive than (for example) 3-clause BSD, but there may be compatible commercial licenses because they are both source-available and permit modification. The implication I can see in some cases is that only one user/entity may be able to view and modify the commercial source code. Of course, SADL itself does not require source to be redistributed except in case of D.2,3.

I apologize for not including the context. This isn't the first release of the license, but it's the first release I plan to eventually submit for review, which is why I omitted mentioning the previous releases. See my reply earlier in this message for links to the prior versions.

On Jan 29 2024, at 1:39 pm, McCoy Smith <mccoy at lexpan.law> wrote:
> > -----Original Message-----
>
> > From: License-discuss <license-discuss-bounces at lists.opensource.org> On
> > Behalf Of Josh Berkus
> > Sent: Monday, January 29, 2024 11:13 AM
> > To: license-discuss at lists.opensource.org; Alec Bloss <abloss at libranext.com>
> > Subject: Re: [License-discuss] Request for Comment: Software and
> > Development License, version 3.0
> >
> >
> > More comments:
> >
> > 1. There's no patent grant in this license.
>
> B.3. is a patent grant.
> >
> > 6. It seems weird to call the first release of a new license "3.0".
>
> Agree with that. Its seems to be an effort to (possibly) indicate that this is an "improved" version over (non-existing?) predecessor versions.
>
> This one has various basic drafting problems, for example the conditions of the license grant are strangely articulated (there's a condition on the copyright grant, a separate condition on all the grants, and no equivalent condition to the one on the copyright grant to the patent grant). I don't see any indication (which is a requirement of submission (https://opensource.org/licenses/review-process/)) that this was reviewed by a lawyer, and it’s not really clear to me that the other requirements for submission of a new license were met either. I think a lot of the drafting issues here could be addressed by having some sort of legal review on this license (and I don’t think submitting it to the list in the hope that the lawyers on the list will do the legal review ought to count).
>
> So I’d suggest further discussion of this license be deferred until the initial requirements of the approval process be met.
See earlier in this reply for context on prior versions. I apologize for not including it initially. It is definitely an effort to improve earlier versions.
I posted here in license-discuss to get input/criticism prior to further revision (which has been received and appreciated), given the amount of licenses that come across the lists, I imagine a lot of things to avoid (or that are problematic) have been seen that may not necessarily be noticed by others. It will be reviewed by legal counsel and revised as necessary prior to submission for review. It is a (very) rough draft but has at least the basic ideas I hope to see in the final.
Thanks!
Alec
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensource.org/pipermail/license-discuss_lists.opensource.org/attachments/20240129/5203382d/attachment.html>