Dear All,
We have discussed the scope note with Carlo, Mark, Patrick and
Christian-Emil
the scope note of E59, but without coming to a conclusion yet. As we are
approaching the meeting, I report the current stage.
One question is, if "Primitive Value" is a good label, given the
extended meaning to complex data types we associate with it.
I propose:
current definition:
E59 Primitive Value
Superclass of: E60 <#_E60_Number> Number
E61 <#_E61_Time_Primitive> Time Primitive
E62 <#_E62_String> String
Scope Note: This class comprises primitive values used as
documentation elements, which are not further elaborated upon within the
model.
As such they are not considered as elements within our universe of
discourse. No specific implementation recommendations are made. It is
recommended that the primitive value system from the implementation
platform be used to substitute for this class and its subclasses.
Examples:
§ ABCDEFG (E62)
§ 3.14 (E60)
§ 0
§ 1921-01-01 (E61)
To become:
E59 Primitive Value
Superclass of: E60 <#_E60_Number> Number
E61 <#_E61_Time_Primitive> Time Primitive
E62 <#_E62_String> String
Scope Note: This class comprises values of primitive data types of
programming languages or database management systems and data types
composed of such values used as documentation elements, as well as their
mathematical abstractions. They are not considered as elements of the
universe of discourse this model aimes at defining and analyzing.
Rather, they play the role of a symbolic interface between the scope of
this model and the world of mathematical and computational
manipulations and the symbolic objects they define and handle. In
particular they comprise lexical forms encoded as "strings" or series of
characters and symbols based on encoding schemes such as UNICODE and
values of datatypes that can be encoded in a lexical form, including
quantitative specifications of time-spans and geometry. They have in
common that instances of E59 Primitive Value define themselves by virtue
of their encoded value, regardless the nature of their mathematical
abstractions. Therefore they must not be represented in an
implementation by a universal identifier associated with a content model
of different identity.
We have not agreed if this additional explanation is useful at all or
more confusing:
"Any implementations of primitive values in a database management system
is necessarily a limited subset of the respective mathematical
abstractions. For instance, the number of character types may be limited
to the UNICODE set, integers to 64 bit representation, or real numbers
to binary floating point numbers with a limited mantissa and exponent.
On the other side, our current scientific ontological understanding of
physical reality regards physical spaces and continuous processes of any
kind to be isomorphic with real numbers or derivatives of the latter.
Similarly, potential symbol sets are unlimited in reality.
Therefore this model, being an ontology and not a data schema, commits
only to the respective abstractions of data types and primitive values,
and not on their implementation-induced limitations. Consequently, any
data schema implementing this model, including all knowledge
representation languages, constitutes a logical subset of this model
with respect to its data values, which may have varying utility in
different application settings. Consequently, no specific implementation
recommendations are made. In a concrete application, it is recommended
that the primitive value system from a chosen implementation platform
and/or data definition language be used to substitute for this class and
its subclasses."
Best,
Martin
--
--------------------------------------------------------------
Dr. Martin Doerr | Vox:+30(2810)391625 |
Research Director | Fax:+30(2810)391638 |
| Email:[email protected] |
|
Center for Cultural Informatics |
Information Systems Laboratory |
Institute of Computer Science |
Foundation for Research and Technology - Hellas (FORTH) |
|
N.Plastira 100, Vassilika Vouton, |
GR70013 Heraklion,Crete,Greece |
|
Web-site:http://www.ics.forth.gr/isl |
--------------------------------------------------------------
