[License-discuss] FAQ entry for slight variations in licenses?

[Engel Nyst]

On 3/7/13, Luis Villa <luis at tieguy.org> wrote:
> A comment on the ISC license page (found by Engel- thanks!) points out
> that there is more than one variation of the ISC license. This is a
> common issue for the older permissive licenses, unfortunately.
>
> Driven by this question, I think we might want a FAQ entry that
> answers the following question:
>
> Q: "I know about a variant of an approved license that differs from
> the approved license by only one or two words. Is the variant also an
> approved license?"
>
> A: Something along the lines of: "Many older licenses have a variety
> of minor variations in the language. Unfortunately, it is not possible
> for OSI to review every variation, so we cannot say if a given
> variation is approved."
>
> Unfortunately, this is a bit of a non-answer, but I'm not seeing a
> good way to address it. Perhaps someone here can give a more
> useful/constructive answer that I'm not seeing right now?
>

I wonder if it needs to be only FAQ (generic answer), or there is
a way to explicitly point out the really minor known variations in
use...

Allow me to ask a different question:
"Is there anything of significance that OSI knows of, at this time,
which would make the variation break OSD in some way, or
otherwise would probably not be approved?"

If the answer 'no', and the variation is in real, successful use,
then perhaps it is worth to say just that.

For example (not necessarily the best example), on the license
page, a note:
"there is a variation in use, this [diff]. While it is not considered the
canonical version approved by OSI, and we do not recommend its
use for new projects, there were no significant reasons brought to
light in community discussions for which this variation would break
the OSD, or otherwise known serious flaws."

It is not approved, but then, no variation is even considered for review
as far as I think I have seen on this mailing list and OSI policies for
non-proliferation or related reasons? (please correct me if wrong).

That doesn't mean that they're all alike...

> Unfortunately, it is not possible for OSI to review every variation,
> so we cannot say if a given variation is approved.

>From "not every", it does not follow logically "not any". I don't doubt
the premise is true, but the second does not follow from it alone.
But perhaps there is no need to go all the way, for minor differences
and existing use...: it is not approved, but sometimes it is discussed
and no reasons are raised, which would probably make it rejected.
(apart from reasons unrelated to its FOSS appropriate status)
With all other disclaimers needed... but if it's true, perhaps there is a
way to say it.

Just an idea... Probably not too well-thought.