Coarray ABI documentation

Coarray ABI documentation

[Damian Rouson]

All,

Please let me know if you contributed to gfortran's Coarray ABI
documentation available at

https://gcc.gnu.org/onlinedocs/gfortran/Coarray-Programming.html

I would like to reproduce this documentation in the OpenCoarrays
source. The gfortran documentation is released under the GNU Free
Documentation License.  OpenCoarrays is released under an OSI-approved
BSD-style license. We will therefore need the approval of each
contributing author in order to relicense the content.  I have already
received the approval of Tobias and Alessandro so I’m just checking
whether there are any other contributors the aforementioned material.

Reproducing the documentation in the OpenCoarrays source will simplify
the process of editing it, increase the likelihood that it will stay
in sync with the source, and allow us to make the function names
compiler-agnostic. Thank you for any consideration you can give this.
If I don’t hear from anyone within a few days, I’ll assume there are
no other authors.

Damian

Re: Coarray ABI documentation

[Janne Blomqvist]

On Mon, Jun 13, 2016 at 8:44 AM, Damian Rouson <damian@rouson.net> wrote:
> All,
>
> Please let me know if you contributed to gfortran's Coarray ABI
> documentation available at
>
> https://gcc.gnu.org/onlinedocs/gfortran/Coarray-Programming.html
>
> I would like to reproduce this documentation in the OpenCoarrays
> source. The gfortran documentation is released under the GNU Free
> Documentation License.  OpenCoarrays is released under an OSI-approved
> BSD-style license. We will therefore need the approval of each
> contributing author in order to relicense the content.  I have already
> received the approval of Tobias and Alessandro so I’m just checking
> whether there are any other contributors the aforementioned material.

I don't recall contributing to this part of the documentation, but in
case I have, please feel free to relicense my contribution(s).

> Reproducing the documentation in the OpenCoarrays source will simplify
> the process of editing it, increase the likelihood that it will stay
> in sync with the source, and allow us to make the function names
> compiler-agnostic. Thank you for any consideration you can give this.
> If I don’t hear from anyone within a few days, I’ll assume there are
> no other authors.

Err, no, I don't this is enough. I'd think you have a hard time
finding a lawyer that would back the interpretation that failure to
respond to a message sent to a mailing list within a few days
constitutes agreement to relicense somebodys contributions. AFAIK what
needs to be done in a situation like this is to dig through the
version control system, and find out everybody who contributed to this
documentation (Which isn't made any simpler by the coarray docs being
part of gfortran.texi rather than a separate file. Too bad!), and then
get an explicit approval from everybody affected. And if somebody
refuses, you need to rewrite that persons contributions.

If you go forward with this, could you submit patches to split the
coarray docs into a separate file, with appropriate licensing headers,
from the main gfortran.texi? It's certainly doable, e.g. the
intrinsics docs lives in intrinsics.texi and is combined into the
final doc when generating. That should make it easier to then keep in
sync in the future (e.g. if you want the "master" document to live
under the opencoarrays project, and copied into gcc sources at
appropriate times).


-- 
Janne Blomqvist

Re: Coarray ABI documentation

[Paul Richard Thomas]

Dear Janne, dear Damian,

I do not think that this is a problem. I am sure that Tobias and
Alessandro were the only contributors, unless Damian did too. Ask
Tobias and Alessandro - I think that they were ploughing a very lonely
furrow there :-)

Cheers

Paul


On 13 June 2016 at 19:43, Janne Blomqvist <blomqvist.janne@gmail.com> wrote:
> On Mon, Jun 13, 2016 at 8:44 AM, Damian Rouson <damian@rouson.net> wrote:
>> All,
>>
>> Please let me know if you contributed to gfortran's Coarray ABI
>> documentation available at
>>
>> https://gcc.gnu.org/onlinedocs/gfortran/Coarray-Programming.html
>>
>> I would like to reproduce this documentation in the OpenCoarrays
>> source. The gfortran documentation is released under the GNU Free
>> Documentation License.  OpenCoarrays is released under an OSI-approved
>> BSD-style license. We will therefore need the approval of each
>> contributing author in order to relicense the content.  I have already
>> received the approval of Tobias and Alessandro so I’m just checking
>> whether there are any other contributors the aforementioned material.
>
> I don't recall contributing to this part of the documentation, but in
> case I have, please feel free to relicense my contribution(s).
>
>> Reproducing the documentation in the OpenCoarrays source will simplify
>> the process of editing it, increase the likelihood that it will stay
>> in sync with the source, and allow us to make the function names
>> compiler-agnostic. Thank you for any consideration you can give this.
>> If I don’t hear from anyone within a few days, I’ll assume there are
>> no other authors.
>
> Err, no, I don't this is enough. I'd think you have a hard time
> finding a lawyer that would back the interpretation that failure to
> respond to a message sent to a mailing list within a few days
> constitutes agreement to relicense somebodys contributions. AFAIK what
> needs to be done in a situation like this is to dig through the
> version control system, and find out everybody who contributed to this
> documentation (Which isn't made any simpler by the coarray docs being
> part of gfortran.texi rather than a separate file. Too bad!), and then
> get an explicit approval from everybody affected. And if somebody
> refuses, you need to rewrite that persons contributions.
>
> If you go forward with this, could you submit patches to split the
> coarray docs into a separate file, with appropriate licensing headers,
> from the main gfortran.texi? It's certainly doable, e.g. the
> intrinsics docs lives in intrinsics.texi and is combined into the
> final doc when generating. That should make it easier to then keep in
> sync in the future (e.g. if you want the "master" document to live
> under the opencoarrays project, and copied into gcc sources at
> appropriate times).
>
>
> --
> Janne Blomqvist



-- 
The difference between genius and stupidity is; genius has its limits.

Albert Einstein