Re: Implementing advanced scientific calculator.

[Sergei Steshenko]

________________________________
From: Sebastian Schöps <address@hidden>
To: address@hidden 
Sent: Friday, March 16, 2018 7:46 PM
Subject: Re: Implementing advanced scientific calculator.



Dear Sergei, 


Sergei Steshenko wrote
> Several years ago Octave package system was conceptually broken. It
> probably still is.
> If so, a probably useful task would to define first and maybe implement a
> new package system.
> Another useful feature would be to write Octave to Julia (
> https://julialang.org/ ) translator.
> Also, there is an interesting library to convert Octave code into C++ :
> https://bitbucket.org/cyrille/cauchy. I haven't tried it yet. If it at
> least somehow works and demands improvement, improving it would be a great
> thing.

the community has already created a website for this kind of recommendations
[1]. For example, it contains a project on the package system wich goes
beyond "it's broken, fix it". So, please add your proposals to [1] if you
like but if you do, then make sure that there is a mentor. By the way, I do
not see why it should be of interest for the Octave community to write a
translator from Octave to Julia?

Sebastian

[1] https://wiki.octave.org/Summer_of_Code_Project_Ideas.




--

Sent from: http://octave.1599824.n4.nabble.com/Octave-General-f1599825.html

---------------------------------------------------------------------------

"I do not see why it should be of interest for the Octave community to write 
atranslator from Octave to Julia?" - for several reasons:

1) Julia is a much better language from the point of view of consistency and 
human understanding, so ultimately one should better switch to Julia and not 
lose the ability to use already developed Matlab/Octave code;

2) Julia is typically faster;
3) it looks like at the moment already there are more packages for Julia than 
for Octave;

4) Julia language runtime is distributed under MIT license - opposed to 
asphyxiating corporatocracy GPL (pretending to be a free license). I.e. end 
user can develop code in Julia and run it in closed source form at the 
place/device of final deployment.

SciLab has Matlab language frontend.

--Sergei.