Re: [boost] different matrix library?

Subject: Re: [boost] different matrix library?
From: joel (joel.falcou_at_[hidden])
Date: 2009-08-15 05:53:07

DE wrote:
> well it seems trivial to me
>     #include <the_lib/lib.h>
>     #include <the_lib/threading.h>  //enabling threading for the_lib
>
> that would be perfect for me
>
> MPI,SIMD ...
>   
Yeah, come back in 2-3 years :) It's all but trivial. nT2 is only 
dealing with SIMD, openMp and threads and is already a 4 years work of 2 
peoples.

>> Will you interface looks like matlab, maple or
>> mathematica ?
>>     
> i prefer to model STL interfaces where appropriate
> in general: as common as possible
>   
Except most user (mathematician) don't even know them. They want to 
build matrix liek they do with their high level prototyping tools.
They don't care about modeling Random Access Sequence or needing to 
fullfill DefaultConstructible concepts. That's why i consider Eigen2
and other attempt as too far from real user concern.

STL like interface in addition to super-easy matlab is better but a dry 
STL complaint will get nobody to use it apart from already hardcore 
developers.

>> That's why I'm avoiding to comment your code cause I'm developing 
>> something similar but for a somehow different audience than yours and my
>> view will prolly be radically different than yours.
>>     
> but i can get some useful stuff from a radically different design and
> utilize it to make the design better
>   
What I mean is that you're starting to duplicate all the thinds I 
already dealt with during my PHD and the last 2 years of research and 
development. But if you like banging heads on walls that other people 
already removed elsewhere , it's up to you. What I wanted to say is that 
it's maybe better if we team up, reusing our existing code base and 
boostify it.

-- 
___________________________________________
Joel Falcou - Assistant Professor
PARALL Team - LRI - Universite Paris Sud XI
Tel : [phone]