[PD] Re: Serious development (was re: Puredance)

[padawan12]

On Wed, 28 Jun 2006 18:29:41 +0000
carmen <_ at whats-your.name> wrote:


> if you don't mind, I'm interested in quoting a few bits that you saw fit
> to put on the webpage, but not mention the list. we all love criticism!!

There's plenty that I see fit to put on the website, but not to mention. Period.
;)

> if there was anything that was a serious roadblock, let's hear it..

Actually, there were no "serious" roadblocks. That's really the point.There were
annoyances that wound me up some, but far less than, say dealing with Red Kens
traffic Stasi.

I think Puredata is becoming everything Miller and the community have been aiming
for. I don't think it's too grandiose to call it a working meta-application.
Much of what Heather and I did for this took a few hours, the rest was refinement.
Most of the grumbles would apply to any system in the same stage of dev as Pd,
duplications, unpublished deprecations, platform specific idiosyncracies,
difficult dependencies to compile certain things.

For example the ghost of [readsf~] still haunts [sfread~], a working [readanysf~]
is still just a rumour, [hid] has been a moving target for months, and for some
reason really basic essentials like [adsr~] and [ead~] seem to get left out of
some builds in favour of far more exotic cruft.  But as a regular Pd user you get
to know the landscape and work around these things.

For the record the reason I try not to post whines up here is 

1) It's not my place to do so until I actively engage in development.
2) From where I'm standing everything seems to be in hand. Finer minds than I are
busily sorting out many of the problems. Hans is bringing up a coherent build proc
so that all platforms will soon be singing from the same hymn sheet. A deprecation
process is evolving so that old code can gracefully die. Miller still meanly keeps
the core clean and small, exactly how it should be. All is well.

That said, we do have a bit of a dining philosophers issue, too many forks and not
enough talking, but that's what you get with a spicy meatball like Pd where theres
lots of vision and enthusiasm. If theres one thing that *fundamentally* annoys me
about Pd (read - it's too late to do anything about it so just shut up about it)
it's the typing, which is plain ugly imho. But nobody ever fell in love with
perfection. 

I think the next big important issues in the life of Pd are things "surrounding
the process". For example, as I said, I'd like to be able to take netlists to useful
source code or linkable object code in an environment where FlTK or GTK GUI toolkits
can be used to wrap up a proper stand-alone application.

I'd like to see  Pd integrated right into the heart of Blender as the "sound engine"
so I can further develop my work on closely coupled physics based synthesis for games
worlds.

I'd like a pony. 

> we really want this thing to be able to scale to a park full of dance pads, one day...

I think you'll eventually see a park full of flying ninja robot monkeys on dance pads
with lasers on their heads. Give it some time. :)

In summary - the task of building an app in Pd was, on the whole possible, quick and
quite pleasurable.

Best,
Andy