Musings on the fundamental nature of data

["Costello, Roger L." <costello@mitre.org>]

Hi Folks,

The subject line is the short title. Here is the longer title:

    Musings on the fundamental nature of the knowledge 
    that is captured from a domain expert and the subsequent 
    exchange of that knowledge as data

Consider this scenario: You conduct multiple in-depth discussions/interviews with someone who is a Subject Matter Expert (SME) and you capture that person's knowledge by encoding it as data using some syntax such as XML. 

What knowledge do you deem important for encoding as data? 

Presumably the data will be shared -- exchanged -- with others. What data do you want others to have?	

This knowledge might be worth encoding as data:

    If there are geometric shapes, then it might be 
    worthwhile to encode knowledge about the shapes 
    so that others can reconstruct the shapes, should 
    that be desired.

    If there is a need to control something, then it might 
    be worthwhile to encode knowledge about how to 
    control the "thing" so that others with access to a 
    "thing" can control it, should that be desired.

    If there are places involved, then it might be worthwhile 
    to encode knowledge about the location of the place so 
    that others can go to that place (or send something to 
    that place), should that be desired.

    If there is an item for which there are many occurrences, 
    then it might be worthwhile to encode knowledge that 
    identifies the item so that others can obtain their own 
    copy of the item, should that be desired.

Observe the recurring pattern. The knowledge that is worth encoding as data is that data which, when exchanged, allows the recipient to replicate, to some extent, the sender's worldview: 

    The sender has a shape. Upon receipt of the shape data 
    the recipient can replicate the shape.

    The sender is controlling some "thing". Upon receipt of 
    the control data the recipient can replicate controlling 
    a "thing".

    The sender is at some location. Upon receipt of the location
    data the recipient can replicate the experience of being
    at the same location.

    The sender has an item. Upon receipt of the identifying
    data the recipient can replicate the item by obtaining his
    own copy of the item.

Assertion: The most fundamental issue with data exchange is: 

    What worldview do you want replicated?

I welcome your insights on this subject.

/Roger