Feynman on the Tree of Life

Where is games' Alan Moore?

all my best work
(imho)
falls in the no man's land
between robot and wisdom
science and literature

and i've learned patience
regarding recognition

(there's a loose inverse correlation
between how highly i rate
any given page on my website
and how many inbound links
it's managed to attract)

but the general topic
of psychological semantics
now grows perceptibly in popularity
year by year
(eg especially with The Sims
and the Semantic Web)

but when an A-list brainiac
like Quinn Norton
asks the very basic question
Where is games' Alan Moore?
it forces me to recalculate
the yawning gap remaining

and seize the opportunity
to sketch my analysis
of gap-bridge engineering...




we're long overdue
(last i checked)
for a website about game-semantics

that reverse-engineers
for every important game
what internal variables it deploys
to map the changes in its
state of play

especially those variables
describing players
and NPCs
(non-player characters)

so most all games know
players' X and Y positions
and probably their state of health

and possibly some of their
strengths and weaknesses

and adding new variables
is one important way
game designers can innovate

because each such variable
allows a small set of new subplots

verbs that increase its value
or decrease its value

powers that emerge when its value
crosses a hidden threshold

and even Alan Moore
could do little or nothing
of enduring literary value

with the impoverished palette
of variables and verbs
so far understood




one of the grandest virtues
in game design
is orthogonality

in which a limited palette
is milked to its utmost
by exhaustive combinatorics

(X-com was outstanding
in this regard)

and it's instructive to explore
the possible plot-combinatorics
of (eg) a familiar variable
like health:

valued-friend-falls-ill
desperate-search-for-power-up-item
friend-recovers or friend-dies

mysterious-plague-infects-many
scientists-seek-vaccine

old-age-brings-health-vulnerabilities

powerful-being-has-achilles-heel




but implementing any of these
in a game
in a dramatically effective way
normally requires
many new supporting variables
which may or may not
contribute an orthogonally
(dis)proportionate
payoff




and when these
foreseeable subplots
are weighed
for literary impact
against, say,
Polti's 36 situations

or any more-recent
inventory of subplots

it becomes clear
that we just don't yet know
where to start
adding literary depth

which variables to choose
to encode which basic set
of stories

More: Representing stories in Web2.0

Values and choice on the yahoo zodiac

if two phrases
have almost identical meanings
they should lie very near each other
in the yahoobet

but

paradoxically

a phrase with exactly
the opposite meaning

must also fall close by

and by implication
all the phrases
for all the values
for a given quality

(eg short medium long, for length)

should also cluster
yahoobetically

(unless perhaps
they're so general
and topic-free
that their position in the yahoobet
is chaotic
and varies dramatically
from second to second...?)




so generally
when we choose between alternatives

setting one to 'green'
and the rest red

all the phrases
denoting all the alternatives
are yahoobetically clustered




and before the choice is made

all might default to yellow




and all philosophy and religion

is about how those yellows

turn red or green

Linktext and heraldic barcodes on the yahoo zodiac

the text of any webpage
can be represented as a star
on the yahoo zodiac

and each possible
sequence of words
chosen to accompany
a link to that page

(ie, the linktext)

can also be represented
as stars




now a visitor to that page

after having evaluated it

could theoretically color 'green'
those examples of possible linktext
that describe what she liked
about the page

red what she didn't like
and yellow if her feelings were mixed

and she could then
theoretically
rank those possibilities
by importance to her
at that moment




so when a web author
wishes to choose linktext
to link a given page

in theory
she should scan those hypothetical rankings
from all her intended audience
top down
to find the highest rated linktext
that's short enough for the context
of her linking page




now if we could analyse
all
the possible linktext rankings
for all possible links

we should notice clusters

that describe variations
in some particular qualities

'long' or 'short' for document length
'lively' or 'dull' for writing quality
'credible' or 'doubtful' for research quality
etc

and these meta-clusters
might be loosely ranked
for how important they generally are

(perhaps with asterisks
specifying conditions
when they'll become specially
important, or not)




the goal of
what i've been calling
heraldic barcodes

is to graphically map
the most important meta-clusters
onto an icon



so, eg document-length
might be a bar along one edge
that's mostly black for short docs
mostly white for long ones

while writing quality might supply
the color-saturation
of the icon's background
grey for dull, bright for lively

etc

with these mappings
chosen by design
to be intuitive-to-remember

and to 'foreground'
the most important values

which alas
are not eternal and unchanging

but rather ever-evolving
in response to changing human needs
and expectations

even allowing
custom mappings
for the custom needs
of a particular group of customers




for the same principles apply
not just to webpage linktext

but to any content-set

eg flickrpix
news stories
news pix
audio files
video files

all of which have
within many different contexts
identifiable value meta-clusters

that cry out for well-designed
heraldic barcodes

Representing stories in Web2.0

something more-or-less interesting
happens

maybe it happens to you
maybe you make it happen
maybe you just witness it

maybe a video camera captures
all or part

so you can post it to YouTube
with a title
and a bit of introduction

possibly when and where
possibly who

some tags that capture
what makes it interesting




if it's not self-explanatory
you may embed it in your blog

if you got audio but not video
you may add photos of the actors

graph their paths in space

if you got photos but
no audio
you may transcribe the dialog

if no photos
you may sketch them
graphically or verbally

maybe as a cartoon
with speech balloons

or as machinima
or posed action figures

or re-enacted by friends
by professionals
by the original players




or you can write a song about it
or weave a novel around it




or you can just retell it in words

typed out
or podcast
or vlogged

tagged maybe
with the actors' names
the props involved
the emotions
the themes and motifs




and before long
one hopes

a scriptable Sims
will let you choose from menus
actors/props/emotions/themes

and the included
storybase
will offer a
default machinima rendering

which you can tweak




but we need to build
that storybase

and it's orders of magnitude
larger
messier
than Sims/Sims2

Sims99, maybe




so to address
that mess

i've proposed this
Tree-of-Life visualisation

as one more
new discipline
for retelling stories

abstracting them down
to branching colored lines

lifelines plus antimath

Joycean epiphanies

the constructions
'had an epiphany' [350k hits]
or 'experienced an epiphany' [25k hits]
occur nowhere in joyce

because for joyce
it wasn't about
discovering something new

it was about
expressing one's true nature

so i find epiphanies
in flickrpix
not when the subject
is somehow wising up

but when they're
showing
their
selves

[more]

Red and green on the Tree of Life

who taught us to dissemble?

to deny our hearts?

for 3 billion years
we did as we liked

but mammon is our master now

dishonor is our code

so back there some ways
some time
in the last 2 million years

our heads got turned
against our hearts

a little red rift
in the lute

that never heals
but somehow spreads

fanned from kith to kin

nursed by a breath unholy

Judgments on the Yahoo zodiac

instead of green for
things you do willingly
and red for unwilling

let green be
things you approve of
and red disapprove

so the teachings of holybooks
can be boiled down to
incomplete and inconsistent constellations
of red and green dots




and consider that
we judge others
by whether they fit
our version of this 'profile'

but that our own
too-human nature
rarely fits it either

and that demagogues
leverage these differences
to incite wars

while most people
would naturally
agree on most values




and consider the revolutionaries
jesus blake nietzsche crowley et al
who reject blind conformity

and nature herself
who knows only natural action
(with bonobos
as the prime incarnation

or housecats
closer to home)

Recap of Yahoo zodiac (plus novelists' styles)

we count the frequency
of the million most-frequent
words on the Web

based on these frequencies
we calculate the expected average distance
between every pair of words

and compare it to the
observed average distance

closer actual pairs
get connected by 'short' elastics

farther pairs by anti-elastics

the whole takes a semi-unique 3D shape




we find the freqency of each word
within each top-level Yahoo
topic category (14 total)

and color it one of 14 colors
to reflect which topic
uses that word most




we rotate the 3D shape
until the 'stars'
of each of the 14 colors
are optimally clustered

and we flatten the shape
along that axis

the radial order of the words
we call 'yahoobetical order'




we temporarily maintain
yahoobetical order
but reposition each word
with nearer orbits for common words
distant orbits for uncommon ones

we now consider word-pairs
on the web
and give them stars
with orbits based on frequency
and positions halfway
between their components

we add triplets, etc
up to phrases, sentences
paragraphs, pages, chapters, books

using hypothetical elastic links
to their components
determining their yahoobetical
positions




we loosen the original
yahoobetical ordering
of the individual words
and let the elastics
re-sort things
(expecting not-too-dramatic
a change)

this is recalibrated
yahoobetical order




we consider the full oeuvre
of any author
and count the word-frequencies

repositioning words' orbits to match

and we compare those
individual-author's orbits
to the whole-web orbits

and re-map based on differences

so the words leCarre (say) uses
more than average
get distinctive orbits
and ditto, those he uses less

(we may choose to filter out
words and phrases used more mainly
in only one or two titles
eg characters' names)

these words' orbits are his
stylistic fingerprint

On a bag of paragraphs

for discussing web content
in the (most) abstract

i favor paragraphs
(rather than sentences
or pages)

since they sometimes scroll off
the bottom of a page
to an archivepage
with a different url

so
many browsing strategies
involve retrieving
sets of paragraphs:

search-engine results
sorted by 'quality'
or by date
or by pagelength

blog-streams
(rss or html)
sequenced by the blogger
followed-in-reverse by the reader


ordinary webpages
with paragraphs sequenced
by the author




and if our browser
views these sets similarly

it can re-sort them
by one anothers' laws

or apply the same filters
(vanish these authors
these keywords
these dates)

Evolving tastes on the Yahoo zodiac

we have the million commonest words
orbiting Moonwise
in yahoobetical order

for a given webpage
the words that don't occur
stay at Moonslength

while the more frequent words
are promoted to inner orbits




phrases rarely occur
more than once per page

sentences and paragraphs
almost never
(rarely more than once
per Web)




a page's wordchoices
may or may not
localise its zodiac topic

but if we re-map all the words
on a given topic
by frequency

visible clustering will emerge




and topics
can be associated with websites

and individuals
can light up favorite topics green
unfavorites red

going thru
more and less experimental
lifephases
when new topics are trialled





and sometimes flipflopping
green for red for green

when an old fave breeds contempt
or a lifechange makes
the unwelcome welcome

(especially
commonly
with the birth of the first child

or after a near-death wakeup call)