RE: [humanmarkup] food for thought

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Well I could use a simpleTypes example and do the 
usual "this is one of those but a little different".
BTW http://www.xml.com/pub/a/2001/08/22/easyschema.html
is a good article schema types.
 
Anyway, I used that example because it is the one 
you had in your example and I don't want to introduce 
complexity at this point, simply to say the tools 
can do it but be sure to know that this kind of 
assertion is a "statement of fact of opinion", not a 
slam dunk "always gets the same result every time". 
In fact, if music did the same thing for everyone 
every time, it would be a boring medium, sort of 
like most web pages. :-)  Some people don't realize 
that the author of "I'd Like To Teach The World 
to Sing in PERFECT Harmony" was J. Goebbels. (No, but 
the point is, it is prophecy or exhortation or entertainment 
until you make it a rule and enforce it, then it is 
governance:  Thee SHALL PolkadotCom!).

The rest is so.  We will have to be very explicit 
that trust, digital sigs and all have to be 
considered, but in a way, that is true of most 
anything so I'm not sure what we have to do for 
the next month or so.  The ideas for trust in 
distributed systems predate TimBL by a very 
long time.  I can't remember but it was probably 
the TQM systems where I ran into that first and 
god knows where before that.

Stable cooperating systems (contract-configurable) depend on it because 
otherwise all the energy dissipates in verification pass. 
The way we used to put it was 

"The thing to remember about Machiavelli is, He Was Fired." 

Maybe out of scope for now.....

The "Building a Better Golem" 
paper is supposed to come out in the next 
MIT Markup Technologies mag, so I won't distribute 
here but everyone can be sure that we've been 
doing rounds over the "we aren't the world rulers" 
topics.  We are hopefully building toolkits then 
it is up to the marketeers and other napoleons 
to see if they rule the world.  I keep Gandhi's 
admonition in mind: "they always fall.  Always."

Again, think about a system where at some point 
you think you or someone else might be confused 
and you can use HumanML as a way to find out where 
and what is the expected remedy.  No free lunch; 
just salami that doesn't kill you with food poisoning.

I'm listening to my midis for pop songs I recorded 
from the JV1080 (it was onsale cheap) that I will use 
as backing tracks at my gigs.   I like guitar kareoke; 
it pays, the band is perfect, and I don't have to share 
the stage with other psychoTropes.  Ahhh... here comes 
the "Won't Get Fooled Again" midi followed by the them 
from Titanic.  Nice contrast that....

Len 
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Sean B. Palmer [mailto:sean@mysterylights.com]
Sent: Friday, August 24, 2001 3:43 PM
To: Bullard, Claude L (Len); humanmarkup-comment@lists.oasis-open.org
Cc: michael_lacy@yahoo.com
Subject: Re: [humanmarkup] food for thought


> The best one might do is specify the emotion, then
> provide a list of resources (say clips, phrases, etc)
> and ASSERT these are related by the named emotion.
> This is interpretive work and the hermaneutics folks
> have a heckuva time on such subjects.

Ah, so we get back to the old context/assertion loop. I'm not sure why
bundling resources into one group and saying "there all give this emotion"
is any better than saying "this one brings up this emotion..." and then
graph merging. The benefit you get from doing it that way is that you don't
have to keep manipulating groups, and you can allow for more discrete data.
Sorry, groups just make me think of hierarchies again... ontologies don't
have to be that way.

> Then everyone argues about that.

I disagree :-)

> It is the authority problem sometimes discussed in
> threads about the usefulness of knowledge bases.

Ooh, guess how many of them I've gotten into? The way around it is to
decentralize, bring in the trust mechanism, and then decentralize the trust
mechanism. No one said it would be easy, but at least it's possible. It's
the old newbie "what if HumanMarkup runs amock, and start publishing bad
data?" question. As Manos mentioned, we define the roots of our particular
ontology tree ("the" ontology tree), but we must let people know we're not
going to charge for it [q.v. TimBL's XML 2000 keynote].

Have there been many previous discussions about trust, digital signatures,
and so on? I think that's a very important aspect of HumanMarkup, because
trust is a very important aspect of privacy. If we're aiming to augment
human communication, then we must work with it, not around it.

> One accepts that kind of assertion or refutes it, but it
> turns on local votes and the extent of time the assertion
> is in effect, so it comes down to popular sentiment
> or as the topic map heads say, just opinions.

Yeah... I call it "context", but that conflicts sometimes with an RDF term.
Semiosis people probably have a word for it too - the concept that the
binding between the symbol and the resource, and the relationships in the
system are interpreted depending on the local situation within a given time
frame. Context.

> CCR and Maria Muldaur?  Really!   So I'm not the
> only PsychoTrope here.  I thought we were a Dead
> culture.

No chance. I'm curently listening to "Oh! Darling", and then I'll probably
put on "Axis: Bold As Love" :-)

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .