Can I remove the bw_... part of the parallel package?

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Can I remove the bw_... part of the parallel package?

Olaf Till-2
All,

(using the help-list to reach more people), would anyone object if I
removed the part of the parallel package with the bw_...() functions?
The bw_-part is what I made only by myself a longer time ago, but the
concept is rather esoteric (using an unreliable cluster for longer
single operations with a self-made interface), I doubt that anyone but
myself ever used it, and I even myself didn't use it anymore for a
considerable time.

I wouldn't even think it's worth depricating it before removal, but
that's OFs policy ...

Olaf

--
public key id EAFE0591, e.g. on x-hkp://pool.sks-keyservers.net

_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave

signature.asc (853 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Can I remove the bw_... part of the parallel package?

Carnë Draug
On Fri, 20 Feb 2015 18:36:54 +0100, Olaf Till wrote

> (using the help-list to reach more people), would anyone object if I
> removed the part of the parallel package with the bw_...() functions?
> The bw_-part is what I made only by myself a longer time ago, but the
> concept is rather esoteric (using an unreliable cluster for longer
> single operations with a self-made interface), I doubt that anyone but
> myself ever used it, and I even myself didn't use it anymore for a
> considerable time.
>
> I wouldn't even think it's worth depricating it before removal, but
> that's OFs policy ...
>

It is not OF policy, it is just highly recommended in any software.
What happens is up to you, the maintainer.  But unless there's any
work involved in maintaining the functions for another cycle as
deprecated (like they need fixes for new Octave versions), why not
keeping them?

Carnë

_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Can I remove the bw_... part of the parallel package?

nrjank
On Fri, Feb 20, 2015 at 6:44 PM, Carnë Draug <[hidden email]> wrote:

> On Fri, 20 Feb 2015 18:36:54 +0100, Olaf Till wrote
>> (using the help-list to reach more people), would anyone object if I
>> removed the part of the parallel package with the bw_...() functions?
>> The bw_-part is what I made only by myself a longer time ago, but the
>> concept is rather esoteric (using an unreliable cluster for longer
>> single operations with a self-made interface), I doubt that anyone but
>> myself ever used it, and I even myself didn't use it anymore for a
>> considerable time.
>>
>> I wouldn't even think it's worth depricating it before removal, but
>> that's OFs policy ...
>>
>
> It is not OF policy, it is just highly recommended in any software.
> What happens is up to you, the maintainer.  But unless there's any
> work involved in maintaining the functions for another cycle as
> deprecated (like they need fixes for new Octave versions), why not
> keeping them?
>
> Carnë
>
> _______________________________________________
> Help-octave mailing list
> [hidden email]
> https://lists.gnu.org/mailman/listinfo/help-octave

The conversation reminds me of this:
http://xkcd.com/1172/

Best to deprecate. On the off chance that one user makes use of that
function, give him a warning before his code breaks.  (think of the
children)

nick j.

_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Can I remove the bw_... part of the parallel package?

rICHARD-2
In reply to this post by Olaf Till-2
On Fri, Feb 20, 2015 at 06:36:54PM +0100, Olaf Till wrote:

> All,
>
> (using the help-list to reach more people), would anyone object if I
> removed the part of the parallel package with the bw_...() functions?
> The bw_-part is what I made only by myself a longer time ago, but the
> concept is rather esoteric (using an unreliable cluster for longer
> single operations with a self-made interface), I doubt that anyone but
> myself ever used it, and I even myself didn't use it anymore for a
> considerable time.
>
> I wouldn't even think it's worth depricating it before removal, but
> that's OFs policy ...
>
Just to be clear, this does not touch pararrayfun and parcellfun, right?

Hmmm ... bw_start.m refers to starting parallel computation in a
beowulf cluster. Would the code be a useful starting point if
intending to move some calculations onto a GPU, and hence worth
keeping?

TIA

Richard A Lough




_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Can I remove the bw_... part of the parallel package?

Olaf Till-2
In reply to this post by nrjank
On Fri, Feb 20, 2015 at 10:38:08PM -0500, Nicholas Jankowski wrote:

> On Fri, Feb 20, 2015 at 6:44 PM, Carnë Draug <[hidden email]> wrote:
> > On Fri, 20 Feb 2015 18:36:54 +0100, Olaf Till wrote
> >> (using the help-list to reach more people), would anyone object if I
> >> removed the part of the parallel package with the bw_...() functions?
> >> The bw_-part is what I made only by myself a longer time ago, but the
> >> concept is rather esoteric (using an unreliable cluster for longer
> >> single operations with a self-made interface), I doubt that anyone but
> >> myself ever used it, and I even myself didn't use it anymore for a
> >> considerable time.
> >>
> >> I wouldn't even think it's worth depricating it before removal, but
> >> that's OFs policy ...
> >>
> >
> > It is not OF policy, it is just highly recommended in any software.
> > What happens is up to you, the maintainer.  But unless there's any
> > work involved in maintaining the functions for another cycle as
> > deprecated (like they need fixes for new Octave versions), why not
> > keeping them?
> >
> > Carnë
>
> The conversation reminds me of this:
> http://xkcd.com/1172/
>
> Best to deprecate. On the off chance that one user makes use of that
> function, give him a warning before his code breaks.  (think of the
> children)
>
> nick j.
I've deprecated these functions now.

Olaf

--
public key id EAFE0591, e.g. on x-hkp://pool.sks-keyservers.net

_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave

signature.asc (853 bytes) Download Attachment