Web: Avatar Agency, A Talking Richard Rorty Bot?

Z & Scott's facebook conversation yesterday - conceiving how independent avatars might conversationally emerge, using Richard Rorty as an example, and vis-a-vis poker bots, Chomsky bot and the Turing test


3:31pmScott
Hey, Z, briefly

How's it going?

3:31pmZ
hey scott

i'm doing well, how are you

3:31pmScott
Well, thanks, too.

3:31pmZ
what are you up to these days

3:32pmScott
Any further thoughts about avatar agency?

3:35pmScott
Lost you

3:35pmZ
whoops

you there?

3:35pmScott
What are you working on?

3:35pmZ
i'm in consulting

telecom media tech

3:37pmScott
Have you finished your BA?

3:37pmZ
yeah i just graduated

probably going to law school next year

3:37pmScott
interesting projects, compared with what you might find optimal?

Congratulations

makes sense

what might you find optimal?

3:38pmZ
hmm tough to say

i dunno, i wanted to try the business thing

and making a bit of money is nice, i guess

and its certainly in an area that i'm interested in

3:39pmScott
there are creative opportunities there, too, especially with IT still

I can imagine

3:39pmZ
yeah... i think i'll probably end up going to law school, do ip law or something

its a good merger of my interests and my skills. i'll probably do dc or local politics for a while too.

3:40pmScott
Z is no longer online. The following was not sent:

I'm exploring creating a World University - MIT OCW and Wikipedia

3:41pmZ
what are you up to these days, scott?

3:43pmScott
you may have two emails in facebook about what I'm up to - teaching "Society and Information Technology in SL, ethnography and world university

1
Teaching "Society and Information Technology" in SL, and working on an ethnography about Harbin Hot Springs, with a 3-D modeling aspect - harbin.org
And you?

2
I'm exploring creating a World University - MIT OCW and Wikipedia - where anyone might post a course or request a course, - and probably with a foundation, as well as opportunities to teach.
http://worlduniversity.wikia.com/wiki/World_University


World University is a very open ended project, and potentially will continue that way

3:45pmZ
gotcha

that's cool stuff

so you must still be very involved with berkman

3:45pmScott
I'd like to start a foundation for World Univ

What's optimal is an interesting question in terms of learning and Universities. What would be most fun ways or subjects to learn for you at the Univ level?

3:47pmZ
hmm

3:47pmScott
I'm in touch with a number of faculty at Berkman - just heard Palfrey talk about "Born Digital" in SF

3:47pmZ
yeah, he's a rockstar.

3:47pmScott
really - optimality that soars, thinking broadly

:):)

3:48pmZ
gosh, i don't know--i feel like i'm pretty biased from my college experience

there were courses that i thought i'd like that i was bored sick of

and courses i thought would be boring that were incredible

intro computer science, for instance

3:48pmScott
and the incredible ones were? And why?

3:49pmZ
i enjoyed cs, intro psych

3:49pmScott
structured flow experiences that led to constructing something?

3:49pmZ
kantian ethics

all sorts of weird stuff

yeah that helped

but ultimately just stuff i was interested in.

3:49pmScott
why kantian ethics?

I can see why the 1st two might be interested in terms of 'flow'

3:50pmZ
it's just fascinating. kant's ethics is sublime--the epitome of pure reason.

3:50pmScott
probably similarly with kantian ethics?

3:50pmZ
kant's ethics is very... analytic

very logically driven

3:50pmScott
is it well structured? or one aspect?

3:51pmZ
well a great professor

well structured by any standard, i'm sure

but its hard to compare across fields like that

3:51pmScott
Yea - Was psych logical, as well? or insightful, or both?

3:51pmZ
more the latter.

3:52pmScott
they are such distinct fields

3:51pmZ
just really fascinating stuff -- this is how we think.

sorry i can't be more helpful. i'll think about it, though.

3:52pmScott
i'm curious, - about psych vis-a-vis avatar agency, too

3:52pmZ
definitely.

3:52pmScott
Rorty recorded into a database and then delivered semantically and interactively through an avatar?

3:54pmScott
if you come across any parts of that scenario - programs, etc. - please let me know - do you know of any?

3:54pmZ
yeah

3:54pmScott
stating with poker bots that might be coded to deliver sentences rather than poker hands

for example

3:55pmZ
that sounds sweet

3:55pmScott
and realizable, - but how?

3:58pmZ
hmm

i imagine that would be the easy part

making the delivery meaningful, harder

4:00pmScott
so the 'semantic' program might get as close as the chomsky bot, for example, but not closer?

4:00pmZ
yeah

at least you'd have to do some convincing

4:01pmScott
more bandwidth? :):)

4:01pmZ
always a start.

4:01pmScott
run chomsky bot faster, and get to more reaslitic meaning more quickly?

believable, that is - closer to Turing's ideal (that's what it is, isn't it?)

4:03pmZ
yeah

will it be enough? i guess we'll see.

4:04pmScott
why not start afresh with a sophisticated poker bot, or does that basic software 'underlie' the chomsky bot?

4:06pmScott
...where the playing of hands was accelerated with bandwidth, such that semantic delivery developed in 3-4 directions, and thus conversationally, and thus depart from the premises of Chomsky bot software?

4:08pmZ
developing the semantic delivery doesn't solve the root problem though

4:08pmScott
a kind of theater-game, word game, dialogue generativity software approach

4:08pmZ
it's still an algorithim

4:08pmScott
for flexible understanding

4:09pmZ
understanding is a malleable term

4:09pmScott
and novel conjunctions of ideas?

4:09pmZ
by understanding, you mean it's a mutating algorithim, or some such thing

the argument isn't that such an algorithim couldn't solve the problem. it's just that its highly unlikely given we don't understand what the problem is.

4:10pmScott
mutating or conversational

4:10pmZ
we don't know enough about our own language, our own brains, to create a machine capable of emulating it

4:10pmScott
defining understanding in computational terms? or defining conversation similarly? or, in terms of Rorty, defining philosophy?

so, semantics, superficially, is what you mean

4:11pmZ
more or less

i'm just a skeptic of reverse engineering.

4:11pmScott
which, I wonder?

4:12pmZ
superficially is too perjorative

4:13pmScott
I have to go soon, but Terrence Deacon at Berkeley who wrote "The Symbolic Species: The Co-Evolution of the Brain and Language" (his Harvard Ph.D.) takes a reverse engineering approach, - there are lots of sophisticated communicative species as a start

4:13pmZ
i'll keep an eye out for it

thanks for the heads up

and for the conversation.

4:13pmScott
I have to do something now, but let's talk soon - likewise

ciao

4:13pmZ
cheers scott

















(http://scott-macleod.blogspot.com/2008/09/web-avatar-agency-talking-richard-rorty.html - September 23, 2008)