Re: HM.VR_AI: Updated

[Rob Nixon <robnixon@execpc.com>]

Subject HM.VR_AI: Goals and Overview : HumanML_VR_AI Facilitator

HM.VR_AI Facilitator: Rob Nixon

-------------------------------

October 5, 2001

-------------------------------
Goals and Overview
-------------------------------
 
Goal 1: To perform an in depth overview of existing standards, common practices, and literature related to Virtual Reality (VR) and Artificial Intelligence (AI) with application to HumanML
 
Goal 2: To establish guidelines and requirements for the use of  Human Markup Language (HumanML) in relation to VR and AI.

Goal 3: To address all existing forms and methodologies related to VR and AI applications of HumanML, including those under development, and make recommendations for standards for development, programming, and collaborations involving VR and AI applications of HumanML
 
Goal 4: To ensure all VR and AI related utilization of HumanML conforms to HumanMarkup Specifications
 
Goal 5: To ensure that HumanML will provide a robust, flexible, and extensible specification for applications involving VR and AI.

Goal 6: To act as common resource and provide support for the VR and AI Communities utilization of HumanML.
 
Action: Facilitator will actively seek the support of the various researchers within the communities to assure that the HumanML specifications will provide the necessary structure for VR and AI application.

--------------
General Overview and Working Hypothesis
--------------

Recognizing that one of the most extensive definitional frameworks for HumanML will be the large base of existing research and development of VR and AI, it is anticipated that immediate developmental efforts will reach the educational, governmental, social and economic arenas through practical applications of VR and AI more quickly than in other areas. To be prepared for this, we are establishing a structured approach to provide for making HumanML readily useful for applications that bridge the “real” and “virtual” worlds.  It should be noted that it is possible to experience a potential imbalance in available attributes when communication occurs between VR and the real world, and it may be necessary to provide a mechanism to adjust the mapping of attributes from the virtual to the real.

Beyond the support for general human parameters, it will be necessary to support the concepts of “Knowledge” (ie. Specified Rules ) and “Experience” (ie. Learned Association Mappings ).   It will also be necessary to support (roles played by individuals) within social interactions of multiple “entities” (human and virtual)

Beyond small scale Chronemics (the timing of delivery of specific information ), HumanML will need to address long term communications of information over long intervals.   It will be necessary to provide a mechanism that will allow for the “evolution” of the information through successive “eras”.  (i.e. original old English -> transliterated to modern English).   This will especially be true of VR simulations.

-------------
Deliverables
--------------
 
11/02/2001 - VR specific HM document ( HM.VR ) – (1st Draft) HumanML VR specification

11/30/2001 - AI specific HM document ( HM.AI ) – (1st Draft) HumanML AI specification

12/28/2001 - VR_AI specific HM document ( HM.VR_AI) –(1st Draft) HumanML VR_AI interaction specification

12/28/2001 - Resource Lists

12/28/2001 - Reference Lists
 
 

"Bullard, Claude L (Len)" wrote:

 I don't understand the following. Please clarify:

Sorry Len,

The paragraph should have read:

-------

To be prepared for this, we are establishing a structured approach to provide for making HumanML readily useful for applications that bridge the “real” and “virtual” worlds.  It should be noted that it is possible to experience a potential imbalance in available attributes when communication occurs between VR and the real world, and it may be necessary to provide a mechanism to adjust the mapping of attributes from the virtual to the real.

-------

This paragraph is addressing the potential problem one could experience with a feed-back-loop type of communication.   Since we have access to ALL information within a VR enviroment ( we are generating the VR afterall) we could easily have a larger number of elements available then we would have access to in the real world (with the current level of technology, until we can access the matrix directly ; ).   We are trying to address the situtation where we would like to map clusters of attributes to a more restricted number of attributes within the real world model.

Sorry, pasted the wrong copy apparently.   Corrected version above.

------
 

Len>Also note that both chronemics and proxemics have scale issues.

Yes, I agree that there are scale issues related to chronemics and proxemics.

LEN> Personal real time and historical time have their analogues...

Are you aware of any modeling languages that have addressed the personal vs. the historic time aspects?

Thanks again Len.

Regards,
Rob