Robot Wisdom auxiliary

[draft]

i'm preparing to do some experiments
in google's new app engine

the 1st very simple: the 'portaliser'
(portaliZer was taken)
will massage a set of favorite links
into a simple page-layout

but the 2nd is much more ambitious
trying to add time-sequencing
to IMDb movie keywords/tags

eg 'Zoolander'

IMDb: main, summary

keywords/tags: Secret Agent, 2000s, Catch Phrase, Multiple Cameos, Spy, Spy Spoof, Frat Pack, Car Chase, Assassination, Brainwashing, Cameo Appearance, Conspiracy, Fame, Fashion Model, Fashion Show, Fashion, Orgy, Stupidity, Exploding Gas Station, Exploitation, Obesity, Sequel, Vanity, Saturday Night Live, Dwarf, Dwarves, Based On Sketch Comedy, Child Labor, Blackface, Bulimia, Catwalk, Erection, Femme Fatale, Malaysia, Masseuse, Reporter, Time Magazine, Directed By Star, Break Dance, Male Model, Character Name In Title


IMDb ontology experiment: [2005] (Modalities, Dramatic themes, Family themes, Sports, Animals, Music and comedy, Historical period, Crime-related, War-related, Horror and scifi, Sex, Animation)

if you take two specific things

and generalise what they have in common

the generalisation will have
far fewer features
than either specific

so when you generalise all humans
into Everyman

Everyman will be vague and blurry
(PKDick's "A Scanner Darkly"
objectifies this into a camouflage-suit)

and creation-myths often reverse this
with structure emerging out of a chaotic void




the a.i. project 'Cyc' attempts
to build a universal ontology
where the highest
most-general
levels
are abstract
blurry and vague

and Ning.com incarnates this paradox
in its universal 'Content Store'
where all content-types are equal
and relations between content-types
are wholly user-defined




but the world is structured
in familiar and reliable ways

we deal with people, places and things

people and things tracing worldlines
thru places and times

so a universal content-store
could embed these reliable structures

'knowing' that all data
can be viewed on the Tree of Life

if interface-design
ever gets a theory

its core must be
ergonomics
ie
calculating
and minimizing
the effort required
by each interface command




managing Web bookmarks
to minimize longterm effort
requires considerable
advance setup-effort

but will repay handsomely
every time you click




i see three major groupings
for bookmarks:

rss-ables
that largely take care of themselves

non-rss regulars
that i group by update-schedule
as toolbar icons

and irregulars
which i'm addressing now




pick a category like
movies music books games

and think of all your favorite
irregular sites
for that category

eg for movies
i use imdb
irregularly
when i want reference-background
on a particular movie or movie-person

rotten tomatoes
when i want an overview
of critics' opinions

script-o-rama
to look for online scripts

one of the spoilers site
when i want a full plot summary

cap
for its measure of offensiveness

robb's oops gallery
for vidcaps of its nude scenes

etc

and each
irregular time
i need one of these

i have to track down the url
maybe just from memory, retyping it
maybe in my bookmarks file
maybe via google
maybe via another site

but ergonomically
all of these methods are
vastly more expensive
than rss or toolbar icons

so suppose
to optimize
we (each) create a custom page
with a normal toolbar icon

that's our irregulars jump-page

that loads instantly
(by minimizing html-junk
and keeping a copy on our local hard drive)

that sorts these reference-sites
by category:
movies music books games
reference firefox html

with each category getting
one line's worth of links:

movies: imdb tomatoes scripts spoilers cap robb's [etc]
music: webjay yahoo lyrics [etc]
books: OlB pGut used complete dy [etc]
games: reviews jay goodexp [etc]
ref: dict thes rhymes wildcard [etc]
firefox: extensions updates [etc]
html: entities [etc]
etc:

so that your top n-hundred
irregular sites
require the same icon-click
followed by a simple second
scan and click?

isn't this ergonomically optimal?




and further
suppose we use Ning
to share and compare
these jump pages

tweaking them from time to time
and re-generating a simple-as-possible page
for download?

poll: would this work for you?

I've done something similar
I've done something better
I use a similar page created by others
My bookmark manager does fine
I'd use this if it were automated (eg at Ning)
I plan to do it, Ning or not
It's too much work
I don't understand it yet

if adolescence is a time
of awkward inarticulacy
then computer science
is still in its adolescence

and computer languages
suffer bad skin
breaking voices
and outbursts of mild insanity




the most appalling thing
about web 2.0
is that you need a virtual phd
to practice it

web 1 only demanded html
but web2 takes css
javascript
flash
rss
etc etc etc

and before Ning
you were largely on your own
trying to wrestle with these

but Ning has created a
learning playground
where your PHP missteps
won't necessitate a reboot

and where the first step
is as easy as clicking
"Clone This App"




as with Wikipedia
this service is CPU-intensive
so once it catches on
the Peter Principle will guarantee
it always runs slow

but for self-education
it can't be beat




the second step
after Cloning
is currently quite steep

because they assume you already
know PHP and CSS

and any cloned app
confronts you with a dozen or more
file fragments
you'll have to plunge into

so it's important to start
as simple as possible
and it's not obvious
which apps are simplest

(try here
or here)

but it's easy to delete missteps
and no doubt much simpler tutorials
are in the works

(and will someone hurry up
and create an equivalent
for generic Flash?)

an informal community is
(or should be)
crystallising

around the shared challenges of the semantic web

and it will behoove us
to consciously, publicly experiment
with the most interesting web2.0
apps and api's (or playgrounds)

del.icio.us for tagging
wikipedia and wikis for collaboration
ning for the new shared vocabulary of
social apps

and googlebase for structuring data

for ggb dares us
to enforce some chosen semantics on
all
our publishing

so we should explore how we might fit
every other web resource into ggb:

wikipedia articles (everyone their own wikipedia?)
imdb entries
blogs w/comments
google-groups threads
webjay playlists

(i'm assuming that google is thinking
since they already cache the whole web
why not privilege a structured subset
by hosting it outright?)