Quantcast

GSoC 2017 Application - Interval Package

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

GSoC 2017 Application - Interval Package

Urathai
Hi!

I'm interested in GSoC for this year, more specifically I'm interested
in improving the interval package. I have filled in my application and
would really appreciate any feedback.

As I mention in the application I'm not sure about which project, on the
interval package, that I want to work on. Any comments on which one
might be the most important or appropriate one would be great. One idea,
as I also mention in the application, I have is to implement a package
for Taylor arithmetic in Octave. I think this would be of great use for
the interval package as well as Octave in general. Though I am not sure
if it is a good idea to start a whole new package for GSoC and I'm not
sure I could find a mentor for that. If you would be interested in this
please let me know.

Application link: http://wiki.octave.org/User:Urathai

Best regards,

Joel Dahne
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: GSoC 2017 Application - Interval Package

siko1056
Urathai wrote
Hi!

I'm interested in GSoC for this year, more specifically I'm interested
in improving the interval package. I have filled in my application and
would really appreciate any feedback.

As I mention in the application I'm not sure about which project, on the
interval package, that I want to work on. Any comments on which one
might be the most important or appropriate one would be great. One idea,
as I also mention in the application, I have is to implement a package
for Taylor arithmetic in Octave. I think this would be of great use for
the interval package as well as Octave in general. Though I am not sure
if it is a good idea to start a whole new package for GSoC and I'm not
sure I could find a mentor for that. If you would be interested in this
please let me know.

Application link: http://wiki.octave.org/User:Urathai

Best regards,

Joel Dahne
Hello Joel Dahne,

What you are willing to work on really depends on you. If you would like to start a Taylor arithmetic package, why don't you already do so? The GSoC program will not blame you for starting before May 30th. Your repository [1] looks like a good start, what makes you stop working on it? We are interested in long term developers, so we would "blame" you for not working on your package after August 21st ;-) The last thing we need, is another abandoned, half-finished, and unmaintained package. Thus no one stops you and if you notice that you need help or ideas, we might be able to give you useful suggestions. And if it comes to GSoC, we can define precise dates and goals to achieve, without having to implement all imaginable functions. But right now I cannot see that you are serious about the package and hardly support it as I don't know much about your interest on it.

In [2] you write a timeline about N-dimensional matrix support, you didn't even mention a word about this. So do you care about this feature? How does this fit in the picture that you want to create Taylor arithmetic package?

Best,
Kai

[1]: https://github.com/Urathai/octave-taylor-POC
[2]: http://wiki.octave.org/User:Urathai
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: GSoC 2017 Application - Interval Package

siko1056
Urathai wrote
siko1056 wrote
Urathai wrote
Hi!

I'm interested in GSoC for this year, more specifically I'm interested
in improving the interval package. I have filled in my application and
would really appreciate any feedback.

As I mention in the application I'm not sure about which project, on the
interval package, that I want to work on. Any comments on which one
might be the most important or appropriate one would be great. One idea,
as I also mention in the application, I have is to implement a package
for Taylor arithmetic in Octave. I think this would be of great use for
the interval package as well as Octave in general. Though I am not sure
if it is a good idea to start a whole new package for GSoC and I'm not
sure I could find a mentor for that. If you would be interested in this
please let me know.

Application link: http://wiki.octave.org/User:Urathai

Best regards,

Joel Dahne
Hello Joel Dahne,

What you are willing to work on really depends on you. If you would like to start a Taylor arithmetic package, why don't you already do so? The GSoC program will not blame you for starting before May 30th. Your repository [1] looks like a good start, what makes you stop working on it? We are interested in long term developers, so we would "blame" you for not working on your package after August 21st ;-) The last thing we need, is another abandoned, half-finished, and unmaintained package. Thus no one stops you and if you notice that you need help or ideas, we might be able to give you useful suggestions. And if it comes to GSoC, we can define precise dates and goals to achieve, without having to implement all imaginable functions. But right now I cannot see that you are serious about the package and hardly support it as I don't know much about your interest on it.

In [2] you write a timeline about N-dimensional matrix support, you didn't even mention a word about this. So do you care about this feature? How does this fit in the picture that you want to create Taylor arithmetic package?

Best,
Kai

[1]: https://github.com/Urathai/octave-taylor-POC
[2]: http://wiki.octave.org/User:Urathai
Hello Kai,

Thank you for the feedback! I will try to answer some of it here and
will also update my application accordingly.

I'm very eager to work on a Taylor package! However I'm new to this kind
of projects and it will take some time for me to get it ready for
release. Working on it as a GSoC project would allow me to spend much
more time on it and probably get it ready for release during the
summer. There are two reasons why I'm not sure about doing this as a
GSoC project: 1. I'm not sure about how I would go for finding a mentor
for a whole new package (it would be tightly connected to the interval
package so maybe Oliver could do this?) 2. I'm not sure about the amount
of work required for creating a new package, it might not be suitable
for GSoC. If this could be handled I would be very happy to work on this
as a GSoC project. Most likely I will continue to work on it anyway, but
will have less time to spend on it.

Continuing on the N-dimensional matrix support for intervals I think
that this is a really important step for the interval package. First of
all I think it would be required for support of higher dimension Taylor
arithmetic in a reasonable way, this alone would be motivation for
me. Also I think that to make the step from non-validated to validated
numerics easier I think it is import that they have similar support for,
for example, matrices.

Working on the interval package, in any way, would allow me to learn
more about this specific package but also how packages in Octave
normally works. I would learn more about the Octave organization and its
infrastructure. This would prepare my for later on managing my own
package.

I hope this answers some of questions and please let me know if you have
any more.

Best regards,
Joel
Dear Joel,

Okay, then stick with the interval package. I think you and Oliver have some ideas about how you start working on the package.

Best,
Kai
Loading...