my intention
with the Tree-of-Life paradigm
is to explore an increasingly rich
semantic field
while keeping it all grounded in a
mathematically precise
visualisation
i plan to use
wikipedia biographies
as a test case
exploring simple vs complex biographical stories
and simple vocabulary vs complex vocabulary
assigning primary colors to the commonest
words (or concepts)
but words are treacherous ground
more quicksilver than concrete
so to clear our palates
i want to explore first
the semantics of music
previously
we considered the subTree of
creatures with beating hearts
and striped each branch with black and white lines
revealing that branch's heartrate
at each moment
but to simplify the visualisation
we might substitute green for 'resting' heartrate
and yellow for slight acceleration
red for racing hearts
so a mild startle
appears as a yellowgreen blip
on an otherwise-smooth green line
and a full day of jangling yellow and red
is more than any body should have to bear
we also brought in virtual studio musicians
to jam a musical accompaniment
to any given branch's blips and jangles
which accompaniment
could be graphed on the branch
as sound vibrations
painted black to white and back
but this encoding
is even less satisfactory than the heartbeat stripes
because it can't be abstracted
into red-yellow-green
perhaps instead
to capture the changing emotional tone
we could try animating abstract colored shapes
(we've all seen
semi-successful
approximations of this)
so that when we tap into
a random point on a random branch
we can choose audio improv
or abstract cartoon
and we might then try to classify
the changes of shape and color
for every possible lifestory on the Tree
but music and animation will both derive
from the arbitrary choice of instruments,
of playing styles, of visual elements, of animation styles
while the representation we need
must instead be universal
a music without instruments
animation without any predefined paintbox
so our classification of story-changes
has to look beyond
the arbitrary palette
and ask
for each subsequence
of arbitrary sights or sounds
what emotions does it capture
and which does it deny
so we're now inside the skins
of the improvising accompanists
choosing from all the possible
next actions
the fittest one
which is what composers do
focusing perhaps
on their present lifestory
or something past
or projecting themselves
into another's past or present shoes
with Thelonious Monk the archetype
of farfetched innovation
deliberately expanding the known palette
to articulate moods never before spoken (or sung)
rejecting as cliched
every pattern used even once
so that if we endow
our imaginary accompanists
with Monklike superpower
the score they'll produce
could be beyond everything ever heard
no matter whose branch we tap
(remember Joyce chose an ordinary everyman
in Leopold Bloom
to immortalise with a verbal 'score'
of streaming consciousness
beyond anything yet read)
and whatever cliched reduction we can imagine
constrained by existing instruments and conventions
is surely an offense against the Muse
aka Truth
who senses the infinite detail
of every fluctuation