Re: GPU Engine for Octave

[Jaroslav Hajek]

On Tue, Dec 8, 2009 at 9:39 PM, John W. Eaton <address@hidden> wrote:
> On  8-Dec-2009, Jaroslav Hajek wrote:
>
> | I thought there was a reference implementation, but now I can't find
> | it, so maybe I was wrong. In that case, of course, OpenCL libraries
> | would not fall under the System Library exception, because it
> | explicitly says that a free implementation must be available. But I
> | suppose if OpenCL will become widely adopted, a free implementation is
> | a matter of time. Note that the reference implementation would
> | probably not need to utilize GPU at all.
>
> The GPLv3 also provides the following definition:
>
>    A "Standard Interface" means an interface that either is an official
>  standard defined by a recognized standards body, or, in the case of
>  interfaces specified for a particular programming language, one that
>  is widely used among developers working in that language.
>
> so I'm not sure that OpenCL qualifies as a "Standard Interface" for
> the purposes of the GPL.  Is the Khronos Group (the industry
> consortium behind OpenCL) a recognized standards body?
>

And what does "recognized" mean? :D
AFAIK, Khronos also publishes and maintains OpenGL and a couple of
other standards, so I suppose it qualifies.

> Perhaps it is, but my point is that the GPL does not allow someone to
> avoid the terms of the GPL simply by publishing an "open standard" and
> then writing some halfway working "reference implementation".
>

"halfway working" would surely not be enough, but otherwise this seems
allowed if the standard satisfies the GPL demands.

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