Hi
I'm sorry about the late reply. I just don't have the needed time at the
moment :-(
man, 01 11 2010 kl. 10:07 +0100, skrev Olaf Till:
> there was some polemic some time ago making me trying to keep Matlab
> compatibility of leasqr while introducing new features
> (constraints). Since until today I had not installed Matlab, I relied
> on feedback if some compatibility matter escaped me.
I feel your pain. Make code that is supposed to be compatible with
software you don't have access to is just a pain!
> There has been no such feedback. But today I noticed by chance that
> Matlab does not allow variable names shadowing functions. I had used
> this extensively for anonymous functions in leasqr. Changing this does
> not simply mean renaming of variables, but would involve changing a
> concept.
>
> So
>
> 1. although leasqr must by mistake have been Matlab incompatible, this
> does not seem to have disturbed anybody,
>
> 2. it is difficult for me to maintain Matlab compatibility, especially
> if there is no feedback,
>
> 3. Matlab incompatibility was not needlessly introduced, but was the
> result of new functionality.
>
> Taken together, I would tend to drop all efforts to keep Matlab
> compatibility of leasqr.
I guess my opinion is formed from two thoughts:
1. In the end, the person who does the programming gets to decide.
2. While we can't ensure perfect compatibility without help from
the user community, I don't see why we shouldn't fix
compatibility issues when they are reported.
So, I guess my opinion is that we should strive to be compatible, but
that nobody should be loosing any sleep over potential non-reported
incompatibilities.
Søren
------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev
