On Thu, 2008-03-06 at 12:28 +0100, Kassen wrote:
> alex said:
> Perhaps something like max/puredata. These graph based
> languages still
> have textual labels like other languages, but the syntax is
> done by
> connecting the labels together with wires, to particular
> inlets and
> outlets.
I think I was misunderstood here. I wasn't saying that max/puredata
don't have syntax, but the opposite. I was saying that they _do_ have
syntax, but asking how could we replace that syntax of wires with a
semantics of space?
> As I see it most graphical systems for programing (at least the ones I
> know) enforce syntax by simply not allowing you to do things that
> aren't according to the syntax.
Yes, but as an aside I note that this is not perfect in PD. You have to
name a delay, and a delay with that name can only exist once, and if you
copy and paste it you end up with a non syntactically correct graph, and
have to go and edit the tcl code to fix it. Well in the version of PD I
have installed anyway.
> Perhaps I'm missing a deeper layer of what you are hinting at but I
> think they are dealing with the syntax for you, it's still there. This
> is often very convenient and at times a bit like Orwell's "newspeak".
Yes I agree, the syntax is there, but I'm wondering if you could get rid
of it and replace it with semantic distance relationships. That is,
instead of expressing semantics within syntax, express semantics within
spatial dimensions.
> The only way out -to me- seems a system where everything that can be
> expressed is meaningful. If we wanted to nit-pick (which I honestly
> don't!) we'd have to say it comes down to a "system where everything
> that can be expressed within the system is meaningful", and admit
> there is still syntax there. If I take a Reactible, take three of the
> blocks put to the side and start juggling with those that's clearly a
> expression that's outside of the legal range of expressions to the
> Reactable.
Yes, there are still the tell-tale signs of symbolic logic shining
through. But perhaps we could get away from this somehow.
> Perhaps a "system without syntax" would have to be a system without
> bounds or one that keeps growing?
Or one on a non-euclidean surface, perhaps where you could zoom in
infinitely.
> I'd start by re-defining the question. I'd start by trying to focus on
> what's really appealing here, being a system that doesn't limit
> impulsive creativity by having us remember what we can and can't do
> yet offers a range of things that can be expressed that's as large as
> possible given that.
Also a system that doesn't focus us on playing with names of things, but
with the meaning of things.
> I'd also like to add that becoming consciously aware of the syntax of
> graphical systems sounds very healthy to me. Often one hears musicians
> say "with system X you can do anything" and is forced to realise those
> musicians subconsciously limited what they thought was possible at all
> to the range of things offered by that system. Back to
> Newspeak....(hey, that last sentence wasn't syntactically a sentence,
> funny).
:)
alex
Received on Thu Mar 06 2008 - 12:23:55 GMT