[License-review] [3rd Resubmission] ModelGo Zero License (MG0-2.0) AND ModelGo Attribution License (MG-BY-2.0)

[Ruby Anna]

Hi Yuta,
Thank you for taking the time to review the text and dig into the March
2025 history.
You have interpreted the text and our intent perfectly. Your understanding
is 100% correct: hosted-only access does fall under the definition of
"Distribution," but because remote users only receive Output (and not the
Licensed or Derivative Materials), Section 2.2(a) does not trigger notice
obligations for those end-users.
As you noted, removing the Output-notice obligation during the earlier
review was a deliberate choice to ensure compliance with the OSD regarding
mere use. We also agree with your point regarding service-page wording—if
we clarify this further, it will be to outline a path for satisfying
existing attribution requirements rather than inventing new ones.
Thank you again for the clear and thorough analysis.
Best regards,
Tyrel
------------------------------

On Sat, Sep 5, 2026 at 12:31 PM 横井雄太 <yuta.yokoi.r at gmail.com> wrote:

> Moming,
>
> Thank you. That resolves the question I was trying to isolate.
>
> I now understand the intended operation of the current text as follows:
>
>    - hosted, API-based, and web-based access is “Distribution” under the
>    definition; but
>    - Section 2.2(a) does not impose license-copy or notice obligations on
>    a hosted-only provider merely because remote access occurs, where the
>    remote user receives only Output and no Licensed Materials or Derivative
>    Materials.
>
> So the scope of “Distribution” as a defined term and the circumstances in
> which the specific Section 2.2(a) obligations have something to attach to
> are not identical. In the hosted-only case, Output alone does not supply
> the Licensed Materials or Derivative Materials to which those requirements
> apply.
>
> The March 2025 review history also makes the result much clearer. In
> particular, it shows that the absence of an Output-notice obligation in the
> current text is not simply an accidental omission: such a requirement was
> considered and removed because of the concern that it would operate as a
> condition on mere use.
>
> I also understand your service-page point in that same light: if wording
> of that kind is added, it should describe a way to satisfy attribution
> where attribution is otherwise required, rather than create a new
> obligation merely because a user accesses the model remotely.
>
> That gives me a clear operational reading of the current text.
>
> Thank you for the detailed clarification and for pointing me to the
> earlier review history.
>
> Best,
>
> Yuta Yokoi
> _______________________________________________
> The opinions expressed in this email are those of the sender and not
> necessarily those of the Open Source Initiative. Communication from the
> Open Source Initiative will be sent from an opensource.org email address.
>
> License-review mailing list
> License-review at lists.opensource.org
>
> http://lists.opensource.org/mailman/listinfo/license-review_lists.opensource.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensource.org/pipermail/license-review_lists.opensource.org/attachments/20260905/7ec0b93a/attachment-0001.htm>