Re: On being Matlab

[Tom Holroyd (NIH/NIMH) [E]]

John W. Eaton wrote:
> In that case, I think there is no point in being "sort of compatible"
> so I say let's design a better language with cleaner syntax and
> semantics.

The source of my frustration is precisely that people keep changing things. 
Code that used to work suddenly breaks. I think one of Octave's selling points 
should be the fact that it is stable.

> I wasn't aware that Octave was generating a lot of warnings about
> Matlab incompatible code.

warning: potential Matlab compatibility problem: " used as string delimiter
warning: potential Matlab compatibility problem: # used as comment character

I don't think these should be warnings. The potential problem doesn't exist.

The "spew" I am getting is really because of a flaw in the .m code I work with 
(sorry, I got carried away); for example,

   warning('off', 'MATLAB:divideByZero');
   out = sum(tmpin, dim) ./ sum(~isnan(in),dim);
   warning('on', 'MATLAB:divideByZero');

I had them off to start with. So setting them on in his code changes what I 
want for the rest of the code. This is just an example, one of several. Is 
there a way to turn warnings off permanently? (I know, fix the problem! Hmm, 
nobody seems to want to do that.)

--
Tom Holroyd, Ph.D.
We experience the world not as it is, but as we expect it to be.