[PD-dev] GridFlow documentation fundraising.

[David Plans Casal]

On 4 Oct 2005, at 09:24, Mathieu Bouchard wrote:

>> This is indeed what I'm trying to point out. Javascript is 
>> underestimated all
>> the time as a language, and having it in Max is a paradigm shift, 
>> IMHO.
>
> Javascript is AFAIK already used to write a large part of 
> Mozilla/FireFox,
> no?

It is indeed. But most people ignore Javascript as a 90's webby thingy.

> And what qualifies something to be called a paradigm shift?

When that something changes the nature and process of the work people 
do with the framework dramatically. Before, I worked in abstractions 
and tried to find other people's externals that might do what I wanted; 
I tried to study C. Now, I enjoy myself and write ruby objects. Not 
big, not clever, but I'm getting more work done than I ever did in a 
year, in a month.

That's a paradigm shift as far as I'm concerned.

> Connecting to C libraries isn't automatic, but there's a Ruby/DL tool 
> that
> can help. For C++ there is SWIG, which currently can't export towards 
> Pd,
> but can export towards Ruby and Python and such.

SWIG is still hard to use IMO, but it would be great to have a small 
SWIG/DL tutorial for Ruby/Puredata, maybe...

> Pd is a lot different because its messages never have return values, 
> which
> is something SWIG can't deal with. This requires reorganisations and/or
> workarounds. What's a float(*)(float) in C becomes an extra outlet 
> _and_
> an extra inlet in Pd, and that's only if you aren't expected to change 
> the
> pointer in the meanwhile, because that'd mean you're screwed.

Yeah. Wow.

> Ruby and Python have their own interesting libraries, which are not
> available in regular C/C++.

Quite.

Does that mean we agree that it's a good thing? ;-)

d

--
# David Plans Casal

problem.each { |day|
	assert_nil(spoon)
}