RE: [huml] Huml vs. HumanML

[Rex Brooks <rexb@starbourne.com>]

Thanks, Len,

I missed this yesterday. The last sentence sums it up nicely and I 
agree. HumanML is a theory about human communication.

Ciao,
Rex

At 9:13 AM -0600 3/4/03, Bullard, Claude L (Len) wrote:
>FWIW:  I haven't reviewed this paper, but a point to make about 
>applicability: 
>
>o  The focus of the HumanML primary schema design was to
>isolate and categorize aspects of human communication, as process
>and as state of the communicators (internally and externally) and
>the environment in which they communicate.  For example, a semiote
>has internal and external aspects that influence the semiotes ability
>to communication intensionally, and it operates within a semiotic
>environment that also has influence over different semiotes attempting
>to communicate. 
>
>At one point, I posted a diagram that lays out
>these relationships.  This looks like high level stuff, but it
>expresses the important aspect of communicating entities within
>a dynamic environment which directs entities and which entities
>direct.  This is the essence of feedback-mediated evolution.  This
>process of evolution is key to the improvement of the process of
>communication and self-directed adaptation.  Modification of self
>and environment through exchange of signs is THE key idea.  HumanML
>is a tool for improving that process, not a tool for external
>control.
>
>o  The schema categories are relatively independent of each other.
>This enables a derivative schema to derive element types from only
>the categories needed by that schemea and to create relationships
>that express the particular theory of human communication being
>schematized.  The HumanML project has a very keen interest in
>testing and improving the primary categories.
>
>This can enable HumanML language families to be very broad and
>even to conflict among themselves as to the implications of the
>theories so expressed.  However, given the focus on observable
>aspects of communication, it is expected that results of observations
>can be fed back to the schema designers to improve their
>results and over time, some theories will converge.  Psychological
>theories are a good example where in our research, we found that
>the theories diverged signficantly.  In that sense, the creation
>of a HumanML derived schema is the codification of a theory. 
>
>There is an old saying in the markup community:  a DTD is a
>theory about a document.  A HumanML schema is a theory about
>human communication.
>
>len
>
>
>From: Ranjeeth Kumar Thunga [mailto:rkthunga@interposting.com]
>
>Hi Guys,
>
>Just a few comments about the paper--although I'm not looking at the final
>version on FTP (Sylvia's folder?).
>
>1)
>HumanML: formal name of language.
>
>huml (lowercase):  namespace prefix and name of discussion list 
>
>Huml (uppercase):  root element of Primary Base Schema
>
>Rex and I chatted about this Sunday, and we realized that we haven't been
>making clear this distinction in the last several months--which we should
>have.  I guess it's just tunnel vision on our parts.  If you guys don't get
>time to do it, that's absolutely fine, but a suggestion is to change the
>references from huml/HuML to HumanML, just so we are as internally
>consistent as possible.
>
>2)
>Another thing Sylvia I believe--Psychology is certainly a disciple for
>applicability for HumanML--you had exclamation points around it with my name
>I believe? ;)
>
>3)
>A final issue is the formatting for some of the text is appearing off the
>page--it may be just my version of Open Office or something.
>
>The paper/presentation should be really nice!  Definitely looking forward to
>seeing the final version.
>
>----------------------------------------------------------------
>To subscribe or unsubscribe from this elist use the subscription
>manager: <http://lists.oasis-open.org/ob/adm.pl>


-- 
Rex Brooks
Starbourne Communications Design
1361-A Addison, Berkeley, CA 94702 *510-849-2309
http://www.starbourne.com * rexb@starbourne.com


----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>