On 06/11/2014 03:37 PM, Richard Fontana wrote:
On Wed, 11 Jun 2014 15:03:29 -0400
Andrew Engelbrecht <sudoman(a)ninthfloor.org> wrote:
>...
> it could be more clear by saying:
>
>> enhancement of the Covered Code, and/or is a runtime
>> library, ...
>
> (by inserting "is" after "and/or")
>
> if this is a welcome addition, i'd be happy to make a pull request,
> unless you want to add it yourself. :-)
Good point! Please feel free to make a pull request or the equivalent.
Richard
thanks!
i just spotted one or two potential issues in the copyleft-next draft
that i'm not sure how to solve:
"Separate Work" means a work that is separate from and
independent of
particular Covered Code and is not by its nature an extension or
enhancement of the Covered Code, and/or a runtime library, standard
library, or similar component that is used to generate an Object Code
form of Covered Code..
IANAL, so i don't know if there are standards regarding where
parenthesis are implied in a sentence based on commas, etc. it seems
like one could interpret the "Separate Work" section like this:
separate work = \
(independent of covered code and not extension) \
_or_ ([licensed source code] is compiler)
once that is fixed, there is a further complication, which may be a bit
of a stretch. i'm going out on a limb here, so maybe i've missed
something obvious. i'll put this here to communicate my thoughts in case
they're useful:
what if the source code to a compiler is compiled with a compiler also
compiled from the same exact source code? for instance gcc-4.9.0 source
code compiled with gcc-4.9.0, which was compiled with clang-x.y? (the
same issue applies to libraries used in compilers, etc.)
if two copies of the same source code are used as the source of both
gcc-4.9.0 binaries, that may present a loop-hole for claiming that the
source used in the first compilation is a "Separate Work" from the
second and therefore a "Mere Aggregation" and not a "Derivative
Work",
according to the license they are both under.
someone might claim the license doesn't apply to one copy of the source,
which would then apply to all further copies of that particular copy.
one could also further claim that exclusion to be viral toward all other
copies.
to complicate the matter, one could put in a copyleft-next'd or GPL'd
useless_functions.c file in one of the copies, to complexify the above
attacker's argument about the separateness of the works.
it might be hard to argue that two equivalent source archives are under
separate, mutually-affecting license conditions, that the license of one
copy could defeat the license of another copy, or be viral, but this
edge-case perhaps ought to be considered while making this section more
explicit and clear.
to summarize:
1.) source code of compilers may simply be considered "separate works".
2.) even with that fixed, there may still be another loop hole.
-andrew