Re: [xml-dev] Should Subject Matter Experts Determine XML Data Implement

["Jeff Greif" <jgreif@alumni.princeton.edu>]

Roger,

It might be worth investigating the subject of "Knowledge Engineering"
from the heyday of artificial intelligence and expert systems.
There's large body of research literature, including various books.

An anecdote:

When working on an expert system that was to determine the optimum
process and cost for making an automobile part from a set of
manufacturing processes, some description of the part, and economic
constraints such as how many pieces per year would be produced, and
the cost of electricity and labor in various factories around the
world, the technical group encountered a machining expert who said,
"Of course, you can calculate how fast you can machine the part on
each kind of lathe, taking into account the multiple tools cutting and
multiple machining stations simultaneously working, but I can do just
as well by playing a movie in my head of the part being produced, and
counting seconds."  Needless to say, we could not do a direct
translation of this methodology in the expert system, and had to get
some instruction on how to do the calculations instead.  The expert
system, built for a very large European company in the early 1990's,
was successful.

Jeff

On Fri, Oct 3, 2008 at 3:57 PM, Costello, Roger L. <costello@mitre.org> wrote:
>
> Hi Folks,
>
> Let me divide the world into two camps:
>
> 1. Subject Matter Experts (SMEs): these are people that are experts in
> a subject (domain), but are not necessarily expert at the technologies
> employed to implement the subject/domain.
>
> 2. Technology Experts: these are people that are experts at the
> technologies, but are not necessarily expert in the subject matter.
>
> -----------------------------------------------
> Consider the following process to developing an XML data
> implementation:
>
> Step 1: SMEs are interviewed about the domain's data and its
> hierarchical relationships. A "Data Specification" is created which
> captures the domain's data and relationships.
>
> Step 2: Technology experts are then handed the Data Specification to
> implement. They create an XML Schema implementation.
>
> -----------------------------------------------
> Questions about the development of the XML implementation of the Data
> Specification:
>
> 1. Should the technology experts be constrained to doing a 1:1 mapping
> of the Data Specification to the XML Schema?
>
> 2. Or, should the technology experts be at liberty to make alterations
> where they see fit? Alterations include using different names,
> generalizing, and reorganizing.
>
> -----------------------------------------------
> Example: an expert on Books (a Book SME) is interviewed. From the
> interview, a "Book Data Specification" is generated. It's a detailed,
> complete document, containing such things as prose about what a Book is
> and the data that characterizes a Book and the relationships among the
> data. Here's a table which succinctly summarizes the Book domain's data
> and hierarchy:
>
> Book
>    Author ..... String
>    Title ...... String
>    Date ....... Year
>    ISBN ....... Sequence of digit, dashes, and 'x'
>    Publisher .. String
>
> The Book Data Specification is then handed off to a technology expert
> for him to create an XML data implementation; namely, an XML Schema.
>
> In a 1:1 implementation, the schema declares a <Book> element that is
> comprised of <Author>, <Date>, <ISBN>, and <Publisher> elements.
>
>  - Effectively, the SMEs are defining the XML data implementation.
>
> In a non-1:1 implementation, the schema diverges from the Book Data
> Specification. For example, suppose the technology expert makes these
> changes:
>
>  - Instead of using the term "Publisher" he uses the term Pub, and
> thus declares a <Pub> element
>
>  - Instead of single term "Author", he declares multiple terms:
>
>       <Person>
>           <GivenName>...</GivenName>
>           <Surname>...</Surname>
>       </Person>
>
>   - Instead of the term "ISBN" he uses the term GUID (Globally Unique
> Identifier)
>
> -----------------------------------------------
> Should an XML data implementation of a Data Specification be 1:1 with
> the Data Specification? In other words, should SMEs effectively define
> XML data implementations?
>
> Or, should technology experts be at liberty to diverge from a Data
> Specification and thus change things specified by SMEs?
>
> /Roger
>
> _______________________________________________________________________
>
> XML-DEV is a publicly archived, unmoderated list hosted by OASIS
> to support XML implementation and development. To minimize
> spam in the archives, you must subscribe before posting.
>
> [Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
> Or unsubscribe: xml-dev-unsubscribe@lists.xml.org
> subscribe: xml-dev-subscribe@lists.xml.org
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
>
>
>