> Jonathan,


> I struggle a bit to understand what is the end goal of Purr Data:
> as far as I can tell, right now it is a re-implementation of the existing GUI
> in HTML5, which is great.
> Once this process is done (and I see we are very close now), how do you see the > project will evolve? Will it just be about making the GUI more
> portable/customizable or do you foresee changes in the GUI/dsp
> communication protocol as well?

It's essentially the development branch of Pd-l2ork, so it inherits all of
its benefits-- infinite undo, native state-saving, GUI theming, and so on.

As for evolving-- you can have a look at the additions I've made to data
structures to see.  With 15 objects one can make a scalar knob that rotates
between 0 and 360 degrees.  With maybe 5-10 more objects one can have
multiple knobs with changeable symbolic names.  Compare that to > 700
lines of C code in the knob external (plus the occasional tcl error, "can't
create" error, etc.)

There's another possibility which I haven't explored-- that is, you can load
anything created with an HTML5 api into the Purr Data GUI and send messages
to Pd from it.

As far as Pd<->GUI communication, I removed most of the tcl commands from
the C code and replaced it with a parameterized interface.  I'm still leveraging
the same socket-based communication.

-Jonathan

> Best,
> Giulio