Re: "optim needs miscellaneous >= 1.0.10" problem

[Thomas Weber]

On Sun, Apr 01, 2012 at 02:02:46PM -0700, Sergei Steshenko wrote:
> 
> I like your answer. I think Mathworks sales and marketing folks will
> like it even more - an excellent bullet why customers should _not_use
> Octave.
> 
> Because they will spend their valuable time figuring out which package 
> versions are (in)compatible with which Octave versions.

ROFL; I have real world experience with Matlab customer service - the
fact that 2 lines of code were enough to crash Matlab brought me that
honor.

And I've seen enough Matlab code that needed changes for a new Matlab
version to know that API stability is something that is essentially
unknown in Matlab.

> Well, it took me several years to convince Octave developers to use
> bug tracking system instead of just a mailing list.

I don't think your input had any influcence in that matter, one way or
another. 

> I'm wondering hoe many years it will take to convince that Octave and
> packages should be released in conjunction with each other.

I don't think you actually understood what I wrote. The intersection
between octave-forge developers and Octave developers is essentially
zero. You are not talking to the right people - you might as well talk
to the Python developers about the Perl release schedule.

        Thomas