[License-review] Request for approval of the resubmitted new "MGB 1.0" license

[Pamela Chestek]

On 2/24/2025 9:05 PM, Pamela Chestek wrote:
>
> On 2/24/2025 10:15 AM, Carlo Piana wrote:
>>
>>     Given that I believe the intent and actual effect of the language
>>     of AFL & OSL -- as well as the actual effect of the language in
>>     MGB 1.0 under current law -- would be to encompass equivalents,
>>     this submission raises an interesting question on license text vs
>>     license intent and approval of licenses. The patent grant of AFL
>>     & OSL has already been approved, and MGB 1.0 attempts to use that
>>     same language in its own grant.* Specifically, can the parol
>>     evidence of the intent of the licensor in drafting language in a
>>     license be considered in both the decision to approve or
>>     disapprove of a license, or for that matter in interpreting the
>>     license itself later when it is used? I tend to think if a
>>     license submitter submits a license with language that under
>>     prevailing law does not violate the OSB (and which has been
>>     approved in the past as complying with the OSB), but the licensor
>>     in submitting the license says they intend for the license text
>>     to violate the OSD, the intent should play into approval even if
>>     the text itself doesn't effectuate the intent. Which in this
>>     case, as I've said before, the intent to carve out DoE from the
>>     license in my opinion violates OSD and makes this license
>>     non-approvable, even if the language which purportedly
>>     accomplishes that intent has previously been approved.
>>
>>
>> This is a recurrent item of discussion. I submit that in a standard 
>> license, the intent of the drafter bears little to no weight in the 
>> interpretation of a legal instrument that is offered for the use of 
>> the many. It has little bearing also in a situation where it is being 
>> used only by the drafter. In any cases, being a matter of 
>> interpretation, which is heavily impacted by the relevant rules of 
>> the applicable law, it would be disingenuous by anyone to attempt any 
>> such interpretation, especially in the absence of an interested 
>> counterpart. Therefore, we, and the Licensing committee first and 
>> foremost, should interpret the license for what it says and according 
>> to the general understanding of similar wording, if there is a 
>> customary use of it in the trade. But the intent, both pro o against 
>> the OSD, should in principle not be considered, unless -- and this is 
>> a notable exception -- if it is expressly stated that one sought goal 
>> is against the OSD.  Which includes carving out any rights which, 
>> under any legal theory or instrument, is required to fully exploit 
>> the grant and do Open Source. I long time ago have been guilty of 
>> attempting to do so, and I have long since repented of my nefarious 
>> deeds.
>>
>> Despite we concentrate on legal text (maybe with an hammer-nail 
>> attitude), the main purpose of clearing a license is to certify that 
>> the text under scrutiny is a legal instrument that enables the 
>> obstacles coming from of "all rights reserved" default as well as 
>> from other potential obstacles coming from other instruments; so that 
>> the recipient is allowed to use of the freedoms that  need to be 
>> granted. The scope of the licensing approval, in other words, is IMHO 
>> that of making sure that the software **is** Open Source, not that 
>> the license is compliant (the OSD is a means to an end). The software 
>> is Open Source if the distribution artifact grants everything that 
>> legally is required and if it provides everything that technically is 
>> necessary to use, study and modify, make copies and distribute 
>> original or derivatives. Anything that does limit or purports to 
>> limit the Open Source-ness of the software (as opposed to imposing 
>> conditions to foster it) is aiming short of Open Source and should be 
>> rejected because it falls short the most important test.
>>
> I think we agree? I'm less inclined than you to trust the language of 
> the instrument because language is so imperfect. But you make 
> allowances for the case where the intended goal is against the OSD, 
> which seems to be this case. I am wary of giving a license our 
> blessing where the license steward believes they have not granted all 
> rights necessary (even when their interpretation doesn't seem to 
> comport with existing law). I think we have a duty to protect the 
> users who will not have seen this discussion and therefore learned 
> that the licensor thinks that some of its patent rights have not been 
> licensed -- if they knew, they surely would not use the software. I 
> think the reasoning is even more compelling here where the license is 
> meant to be essentially the same as the Apache license, that it it's a 
> redundant license, but for the change to the patent language.
>
Two more thoughts - in the US, this colloquy might be admissible in 
court to determine the meaning of the license. It's possible that a 
court could see that the drafter's intent was not to grant patent rights 
and conclude that it hadn't.

I also think it's a bad precedent for the OSI to approve a license where 
we know the licensor's intent is not to grant all rights necessary. We 
don't have stare decisis, but we strive for consistency and it's always 
challenging to explain inconsistency to license submitters.

Pam

> Pam
>
> Pamela S. Chestek (in my personal capacity)
> Chestek Legal
> PLEASE NOTE OUR NEW MAILING ADDRESS
> 4641 Post St.
> Unit 4316
> El Dorado Hills, CA 95762
> +1 919-800-8033
> pamela at chesteklegal
> www.chesteklegal.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensource.org/pipermail/license-review_lists.opensource.org/attachments/20250224/7cfbc79d/attachment.htm>