I suspect that the conversation is getting snagged on a false dichotomy: enforceability is not in fact a binary characteristic of a license, only of a license in the context of a specific dispute, with specific plaintiff and respondent, in a specific jurisdiction, on specific facts, at a specific phase of a legal process (trial vs. appeal), etc. The temptation to leap from "a license cannot be intrinsically enforceable" to "nothing can be known pre facto about a license's enforceability" is an erroneous one. The temptation to shoehorn into simple to understand discrete categories (particularly binary ones) is strong and often valuable, but also often incorrect, as in this case.

This problem is not unique to F/OSS licensing, it applies to all legal drafting[1]. The reasonable response is not to throw one's hands into the air and declare that because perfection is not possible, therefore no worthwhile outcomes can be achieved. It's to draft diligently.

For OSI I'd suggest that this means that a candidate license could reasonably declined with a recommendation to work with a legal professional to tighten up the drafting, if it is apparent during review that there are substantial omissions of conventional drafting practice. (Obviously, "substantial omissions" *also* doesn't fall neatly into yes/no categories. Subjective judgement is required.)

- Roland

1: It applies to all human creative endeavours in fact, even to programming where sometimes things can't readily be shoehorned into neat categories so some creativity must be employed to produce a useful-if-inelegant solution.

------------------------------------------------------------------------

On 26/10/23 07:47, Nicholas Matthew Neft Weinstock wrote:
That's a really interesting way of looking at it:

In order to be within the OSD, the license needs to be actually capable of 
granting the necessary rights/licenses.

That's certainly true in other contexts.  If the license only grants the right 
to COPY but not MODIFY or DISTRIBUTE, it doesn't grant the rights necessary to 
satisfy the OSD.

The same makes sense in the question of enforceability.  If the license is not 
able to grant the necessary rights because it is unenforceable, how can it 
grant the rights necessary to satisfy the OSD?

On the other hand, the difficulty in this analysis is that some matters of 
enforceability might change over time.  If a provision was enforceable when OSI 
approved the license, and then the law changed and it is no longer enforceable, 
does the license status change?  How would that be identified, determined and 
enforced?  Or if a license was rejected due to lack of enforceability, and then 
the law caught up and the relevant provision is suddenly enforceable, would it 
be eligible for reconsideration?

Maybe there are two sub-categories of "enforceability."  So if a provision is 
so poorly written as to be unenforceable, that's something to consider.  But if a 
provision is written well and it's a question of legality, this should not be considered.

-Nick

-----Original Message-----
From: License-discuss<license-discuss-boun...@lists.opensource.org>  On Behalf 
Of Josh Berkus
Sent: Wednesday, October 25, 2023 1:26 PM
To:license-discuss@lists.opensource.org
Subject: Re: [License-discuss] Evaluating the Enforceability of a License 
Should Not be a Criteria for OSI License Review

WARNING: This email originated from outside of Qualcomm. Please be wary of any 
links or attachments, and do not enable macros.

On 10/7/23 22:25, Patrick Schleizer via License-discuss wrote:
>
> Concerns regarding the enforceability of a license, especially across > different jurisdictions, should not be a determinant in the approval > or rejection of a license. Enforceability can vary significantly > across different regions due to the diverse legal landscapes, making > it a less consistent metric for license review. The issue lies in the > fact that enforceability is largely subjective and lacks an > independent, quick reproducible testing framework

Nothing else in the OSD includes an "independent, quick[ly] reproduceable testing 
framework".  Why should enforceability?

Nor can you separate enforceability from OSD evaluation.  If the license 
complies with the OSD only by means of legal text that is unenforceable in a 
majority of the world, then it doesn't comply with the OSD.  The author's 
intent doesn't matter if they weren't successful.

We also regularly have licenses submitted that aren't written with legal advice 
and are so poorly drafted that they would not be enforceable anywhere at all.  
Do you really think that the OSI should approve these?

--
Josh Berkus


_______________________________________________
The opinions expressed in this email are those of the sender and not 
necessarily those of the Open Source Initiative. Official statements by the 
Open Source Initiative will be sent from an opensource.org email address.

License-discuss mailing list
License-discuss@lists.opensource.org
http://lists.opensource.org/mailman/listinfo/license-discuss_lists.opensource.org

_______________________________________________
The opinions expressed in this email are those of the sender and not 
necessarily those of the Open Source Initiative. Official statements by the 
Open Source Initiative will be sent from an opensource.org email address.

License-discuss mailing list
License-discuss@lists.opensource.org
http://lists.opensource.org/mailman/listinfo/license-discuss_lists.opensource.org

_______________________________________________
The opinions expressed in this email are those of the sender and not 
necessarily those of the Open Source Initiative. Official statements by the 
Open Source Initiative will be sent from an opensource.org email address.

License-discuss mailing list
License-discuss@lists.opensource.org
http://lists.opensource.org/mailman/listinfo/license-discuss_lists.opensource.org

Reply via email to