Dear Vladimir,
Thank you very much for this comment.
I take the point to add a counterexample to compatibility as ISSUE.
Vladimir Ivanov wrote:
---------- Forwarded message ----------
From: Vladimir Ivanov <[email protected]>
Date: 2008/9/6
Subject: Re: [Crm-sig] ISSUE: CRM compatibility, VERY IMPORTANT
To: martin <[email protected]>
Dear Martin,
A paragraph is about "without loss of meaning" is very interesting:
"In the context of this chapter, the expression "without loss of
meaning with respect to the CRM concepts" means the following: The CRM
concepts are used to classify items of discourse and their
relationships. By virtue of this classification, data can be
understood as propositions of a kind declared by the CRM about real
world facts, such as "Object x. forms part of: Object y". In case the
encoding, i.e., the language of describing a fact, is changed, only
the expert knowledgeable about both languages can assess if the two
propositions describe the same fact. If that is the case, then there
is no loss of meaning with respect to the senses that can be
classified by CRM concepts. Communities of practice requiring fewer
concepts than the CRM declares may restrict CRM compatibility with
respect to a reasonable, explicitly declared subset of the CRM. "
Few thoughts about it:
The notion of compatibility is based on interpretation of meaning,
which in turn is based on comparing extensions (sets of individuals,
relationships).
And interpretation of meaning is an activity carried out by expert.
It is hard to imagine that an expert will asses extensions
rather than intensions to make a decision.
This is more real than you may believe. In all our data transformation jobs
we were forced to go down to the extensions, i.e. representative samples, to
verify the mappings. Typically, you need an expert user of a database that helps
spotting all unorthodox uses and abuses of database fields. Without seeing any
example, most even elaborate data type definitions stay ambiguous.
Should we distinguish notions of intensional/extensional meaning?
Should we introduce relationships that preserve meaning (equivalence,
subsumtion)?
Yes and no.
No: The idea is that this text is enough to be elaborated outside
of the ISO process as an interpretation of the standard. If we have said enough
in the
standard to do so, we need not say more in the standard.
Then, yes, we need a detailed procedure.
CRM compatibility of data structure depends on declaration
(one may restrict CRM compatibility on one hand and reduce data
structure on another hand).
If there are no incompatible structures, then we could write
"any data structure could be declared as CRM compatible",
else we could provide at least one example of incompatible
structure (with respect to any declaration).
I would say at least one example for each use case declaration.
One data structure-import and export-incompatible example should be:
Physical Object. has date: Time-Span
There is no common abstraction in the CRM to all possible interpretations.
You can import into a CRM compatible form but not transform it back.
Any comments?
To be discussed in Athens.
Martin
Best regards,
Vladimir
2008/9/2 martin <[email protected]>:
Dear All,
Attached my draft trying to flesh out in a text form the decisions made
about CRM comaptibility in the last CRM-SIG meeting in May. This should
become
the instrument to distinguish false compatibility claims. It should be
rich enough to cover all reasonable application cases. If not, we have to
change
it.
Please read carefully, with respect to:
a) Is this clear enough so that a good practice of validating CRM
compatibility
can be based on this
b) Is it logically consistent
c) Is it good English
d) Is it functional and useful, is it attractive for providers and users,
does it foster what we want to achieve with the CRM?
Please comment. This text, your comments or rewritings will be the base for
a
decision on the CIDOC Conference if this text will be proposed to ISO
as amendment to ISO21127. Only proposals submitted to this mailing list
before the meeting will be discussed.
Best,
Martin
--
--------------------------------------------------------------
Dr. Martin Doerr | Vox:+30(2810)391625 |
Principle Researcher | 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) |
|
Vassilika Vouton,P.O.Box1385,GR71110 Heraklion,Crete,Greece |
|
Web-site: http://www.ics.forth.gr/isl |
--------------------------------------------------------------
_______________________________________________
Crm-sig mailing list
[email protected]
http://lists.ics.forth.gr/mailman/listinfo/crm-sig
_______________________________________________
Crm-sig mailing list
[email protected]
http://lists.ics.forth.gr/mailman/listinfo/crm-sig
--
--------------------------------------------------------------
Dr. Martin Doerr | Vox:+30(2810)391625 |
Principle Researcher | 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) |
|
Vassilika Vouton,P.O.Box1385,GR71110 Heraklion,Crete,Greece |
|
Web-site: http://www.ics.forth.gr/isl |
--------------------------------------------------------------
