[License-discuss] Copyright on APIs

[Richard Fontana]

On Tue, Jul 2, 2019 at 2:38 PM VanL <van.lindberg at gmail.com> wrote:
>
>
>
> On Tue, Jul 2, 2019 at 1:32 PM Richard Fontana <rfontana at redhat.com> wrote:
>>
>> On Tue, Jul 2, 2019 at 2:19 PM VanL <van.lindberg at gmail.com> wrote:
>>
>> > On Tue, Jul 2, 2019 at 1:05 PM Richard Fontana <rfontana at redhat.com> wrote:
>> >>
>> >> I believe one could reasonably argue that a reimplementation of an API
>> >> (necessarily copying the supposed expressive elements of the API) does
>> >> not fit this definition (even if it would fit the folk notion of
>> >> derivative work that the free software/open source community largely
>> >> seems to subscribe to).
>> >
>> >
>> > Quick clarification:
>> >
>> > Did you mean that "one could reasonably argue that a reimplementation... fits this definition" but that the folk notion rejects it?
>>
>> No.
>>
>> > Or did you mean that a reimplementation does not fit this definition" (as you wrote it) but that the folk notion accepts that a reimplementation is a derivative work?
>>
>> Sort of: I can see how the folk notion of derivative work (which
>> evolved under the assumption that APIs were not copyrightable, largely
>> by people opposed to copyrightability of software interfaces as a
>> policy matter) could lead one to the conclusion that the
>> reimplementation of an API is a derivative work of the API. Though I
>> haven't really seen any indication that anyone subscribing to that
>> folk notion is actually seriously concerned about this being a
>> consequence of _Oracle v. Google_.
>
>
>
> Thank you. FWIW, my sense would have been opposite: The language allows the interpretation that reimplementation of APIs created a derivative work, but the folk notion was that APIs were excluded. The folk notion was also broadly accepted by legal folks because there was 102(b) that they could point to as excluding APIs. It was due to the violation of the folk notion that OvG is such a big deal.

There was a separate belief (long held not just by the free
software/open source community but by the entire software industry and
developer community, essentially, other than perhaps the legal
department of Sun Microsystems) that APIs were outside the scope of
copyright. So the notion of a derivative work of an API never really
came up.

Richard