2009/7/21 Kassen <signal.automatique_at_gmail.com>:
> The very best we may be able to hope for is complaints about a lack of
> unit-tests, maybe a Dijkstra quote delivered with a frown.
:D
> Perhaps we should
> try entering "battle of the bands" style evenings or show up at nights of
> open poetry?
Yes, there much be plenty of guitar dominated 'open mic' evenings
around that we can inflitrate.
> Maybe that strategy should be combined with a Panasonic
> Toughbook sponsorship.
heh
>> I understand that the throat has evolved to support speech but
>> research into sign languages and their users can readily discount any
>> such strong claims. In particular I think the idea of a speech
>> 'module' tied to articulations of the vocal tract is nonsense.
>
> I would like to point out that I never made that last claim or any claim
> about the actual exact functioning of the brain.
Sorry if I misrepresent you, you mentioned that some think that the
evolution of the throat allowed for evolution of intellectual
abilities. I thought you were referencing something like the 'speech
modules' within the motor theory of speech perception.
> Do you have a link to a
> article "readily discounting" that the development of language led to (more
> advanced) symbol manipulation in human behaviour/thinking and that bonobos
> may be affected in similar ways by teaching them language?
No it was tying such theories to the vocal tract that I was objecting to.
>> Are letters of the alphabet abstract images?
> Not in the sense that I meant here. I'd say they are symbols. I would say
> that the term [SinOsc] (including the letters used in it) conveys meaning in
> a very different way than a Mondriaan painting (I'm excluding his early work
> here).
Yes agreed but there is something funny going on here. Are symbols
more or less abstract than a Mondriaan painting?
> No, the description of that structure may be shorter than a specific
> implementation (it might be longer as well). I meant the more conceptual one
> is a more generalised one, indicating a (infinitely large) set of melodic
> structures.
So a concept is a class of things, I agree. I like the idea of a
concept always being shorter than its instances though, even if it was
a misunderstanding.
> I might be interested in the effect of that type of structure
> itself more than I am in the actual notes contained in any one specific
> example of it.
Yes! This is why generative music that endlessly generates new
variations tends to sound static.
> Livecoding, unlike typical sequencers, allows me to describe
> that type of structure directly instead of indicate it by example.
So one possibility is that we think in symbols, and that the
programming language code is a translation of mentalese to ChucK (or
whatever), where mentalese is some internal human language to which we
have no direct conscious access. The translation is never perfect,
but can be successful to a high degree.
Another possibility is that you after learning the ChucK language, you
think in it, and the code is a transcription of that process.
A third possibility is that we think in terms of a geometry in a
perceived world of movements in spaces, that symbols are just
signposts within that world and language falls out of movements
between learned arrangements of signposts. We can construct ChucK
sourcecode using such movements which then generate music which we
then perceive in terms of movements and shapes in spaces, which the
arrangement of signposts somehow accords with. The concepts are
natively the shapes in the spaces rather than the symbolic signposts
though.
I feel a bit stupid suggesting this wacky third possibility but I
think there could be some truth in it. I've been tempted to start
drawing ascii boxes to make my point clear but no-one ever replies to
a post with ascii boxes in it.
> For example; Wolfgang Voight released a series of records exploring the
> edges of techno (this is a bit of a central theme to his work). One of his
> aliasses is "m:1:5" meaning Maßstab (or ratio) one to five. These records
> hold techno tracks where one rhythm is overlaid with itself at a fifth of
> the speed.
The way I think that I think about a concept like 1:5 is of spatial
relationship rather than language. Different sized rulers lining up,
different sized cogs etc.
> I might agree with that. For safety's sake I should because it's becoming
> increasingly clear that I'm not completely sure what the word "concept"
> means. Can -for example and your opinion- the meaning of the word "concept"
> be expressed in natural language? In a formal one?
We don't yet have a conclusive single idea of what a concept is (how
it is learned and represented) apart from "a mental representation of
a class of things", but once we do I don't see why it can't be
expressed in either a natural or formal language. It'll just be
another human process to observe and describe.
> I am still maintaining
> that I hold that descriptions of musical phenomena in code are often "more
> conceptual" than any particular example of them in played or recorded music
> though.
Ok I agree, it could definitely be closer to a concept in terms of
levels of abstraction, and may even simulate the concept very
accurately.
> As a side note; English has the expression "arguing semantics" which is
> generally used to mean arguing about trivial matters. I still find that very
> odd as I feel attempting to agree on semantics is quite fundamental to
> agreeing in any spoken or written conversation.
Yes!
alex
--
http://yaxu.org/
Received on Wed Jul 22 2009 - 16:45:51 BST