ga octave package maintenance

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

ga octave package maintenance

John Donoghue-3

The ga hasn't been updated since 2012, but looking at bug reports hasn't
really had much of anything wrong until now, where it no longer runs in
Octave 4.4.

Is there still an active maintainer for it ?




Reply | Threaded
Open this post in threaded view
|

Re: ga octave package maintenance

edmund ronald
Much of numerical or semi-numerical code out there is historical. 
It might make more sense to figure out why new releases of Octave break things than to hope to force maintainers to track code changes.

Edmund

On Tue, Jul 24, 2018 at 6:57 PM, JohnD <[hidden email]> wrote:

The ga hasn't been updated since 2012, but looking at bug reports hasn't
really had much of anything wrong until now, where it no longer runs in
Octave 4.4.

Is there still an active maintainer for it ?





Reply | Threaded
Open this post in threaded view
|

RE: ga octave package maintenance

John Donoghue-3


From: edmund ronald [mailto:[hidden email]]
Sent: Tuesday, July 24, 2018 1:41 PM
To: JohnD
Cc: octave-maintainers
Subject: Re: ga octave package maintenance

Much of numerical or semi-numerical code out there is historical.
It might make more sense to figure out why new releases of Octave break things than to hope to force maintainers to track code changes.

Edmund


------

Well in the case of ga, it was code that was returning a value that was never being set. Older versions of octave didn’t care. Newer versions do.



Reply | Threaded
Open this post in threaded view
|

Re: ga octave package maintenance

edmund ronald
Can't we do it like the C guys do, and have a compatibility flag? 

Or even better a line at the head of each file that says which version the author is using, and so subsequent releases know which flags they need to assume when loading that file?


Edmund

On Tue, Jul 24, 2018 at 8:36 PM, JohnD <[hidden email]> wrote:


From: edmund ronald [mailto:[hidden email]]
Sent: Tuesday, July 24, 2018 1:41 PM
To: JohnD
Cc: octave-maintainers
Subject: Re: ga octave package maintenance

Much of numerical or semi-numerical code out there is historical.
It might make more sense to figure out why new releases of Octave break things than to hope to force maintainers to track code changes.

Edmund


------

Well in the case of ga, it was code that was returning a value that was never being set. Older versions of octave didn’t care. Newer versions do.



Reply | Threaded
Open this post in threaded view
|

Re: ga octave package maintenance

Juan Pablo Carbajal-2
If the package is not working and not being actively maintained is
shouldn't it be archived and removed form the packages installable
from octave forge?

Reply | Threaded
Open this post in threaded view
|

RE: ga octave package maintenance

John Donoghue-3


> -----Original Message-----
> From: Juan Pablo Carbajal [mailto:[hidden email]]
> Sent: Wednesday, July 25, 2018 10:27 AM
> To: edmund ronald
> Cc: JohnD; Maintainers GNU Octave
> Subject: Re: ga octave package maintenance
>
> If the package is not working and not being actively maintained is
> shouldn't it be archived and removed form the packages installable
> from octave forge?


Im OK with maintaining it as in applying bug fixes, verifying it will install etc, which was my initial question.


Another question, and probably should be a new email subject is how does anyone know what is maintained or not?

Date last released? The ga package was last released 2012. But given that is has been installing OK until the 4.4 release and didn’t have any major issues, was it maintained or unmaintained?

There are also a number of other packages that are maintained but have yet to have new releases so that they will work for 4.4, so maintained or unmaintained ?

How much is the package used or is it not used by anyone, should perhaps also be a factor on whether a package is just let sink into the archive oblivion or should be picked up by someone if possible, but can we tell that ?