gnucap-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnucap-devel] [GNUCAP][GSOC] Replacement of gwave


From: al davis
Subject: Re: [Gnucap-devel] [GNUCAP][GSOC] Replacement of gwave
Date: Thu, 6 Mar 2014 15:15:56 -0500
User-agent: KMail/1.13.7 (Linux/3.2.0-4-amd64; KDE/4.8.4; x86_64; ; )

On Tuesday 04 March 2014, Abhinav singh wrote:
> Thanks for response. I have prior knowledge related to GTK+
> libraries and Python. So i wants to know what is the main
> focus of this project whether better GUI or we should
> emphasize upon more commands and plugins with simple GUI.

What is a "better GUI"?  I really don't know how to define that.  
Of course we want it "better" but what does that mean?

Going back to the suggested project, "replacement of gwave" ..

Remember high quality code is modular.  A high quality program 
consists of a collection of high quality modules, which consist 
of ...

High quality implied well designed with a clean interface, 
designed to be maintainable, testable, ...  and it should be 
thouroughly tested.  Whatever approach you take, we expect high 
quality.

Now defining the goal ..  First it needs to be something people 
want to use, something that makes gnucap useful to more people.

Python has the advantage of being an interpreted language, which 
brings great opportunity for later expansion.  C++ is compiled, 
and can link to existing gnucap libraries.  By doing so, you 
have the use of gnucap's i/o, the expression handlers which can 
do math on arrays as if they were scalars.  Also you get the 
gnucap plugin system, which can be extended.

A stand-alone program can be useful beyond gnucap.  A plugin 
provide a tighter user experience, for example showing waveforms 
as they are computed.  I think this is a big advantage.

With modular construction you can have both.  The stand-alone 
program can use the gnucap core library and plugin system, just 
a different main.  I think this is important.

Whatever features you leave out, make sure the plans allow for 
them to happen in the future.

As a test, now, before submitting your proposal, try writing a 
simple program that plots data from a file, just as a study, 
just to see how it is done.  It's a program nobody else will 
ever use, but it will give you the insight to take on a bigger 
project.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]