Re: GPU Engine for Octave

[Jaroslav Hajek]

On Tue, Dec 8, 2009 at 10:15 PM, Judd Storrs <address@hidden> wrote:
> On Tue, Dec 8, 2009 at 3:53 PM, John W. Eaton <address@hidden> wrote:
>> On  8-Dec-2009, Jaroslav Hajek wrote:
>> | "halfway working" would surely not be enough, but otherwise this seems
>> | allowed if the standard satisfies the GPL demands.
>>
>> Yes, I agree.  My point is just that some care needs to be taken when
>> evaluating whether a given case is OK or not.
>
> I agree with this, but the GPL is a little murky here. I think the
> intent is to allow a proprietary library to substitute for an existing
> free-library if and only if the GPL-covered program and the library
> only interact via the "standard interface". There is maybe an
> ambiguity in determining what the requirements of the publicly
> available version are. i.e. does "available to the public" mean
> GPL-compatible? From GPLv3 section 1:
>
> The “System Libraries” of an executable work include anything, other
> than the work as a whole, that ... (b) serves ... to implement a
> Standard Interface for which an implementation is available to the
> public in source code form.
>
> The ambiguity is what does "available in source code" form entail? The
> bare minimum would be code that you are allowed to read. Are they
> releasing the publicly available version of the library from full
> GPL-compatibility? Or even from being free-software at all?
>
>

I believe this means a free software or public domain implementation,
but not necessarily GPL'ed.

-- 
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz