[Crm-sig] Conceptual Objects and their Appellations, ISSUE
martin at ics.forth.gr
Sun Sep 6 22:27:27 EEST 2015
Indeed, the specializations of E41 Appellation by category of named item
are a source of confusion. I'd rather vote to abandon them!
The point is, that the CRM must not contain classes which are a pure
deduction of relationships. The fact, that an appellation is
occasionally used to identify a Conceptual Object does not tell anything
about its ontological nature. In the initial design, we had in mind
research cases, in which we find appellations which, by their form,
exhibit to be designed for a certain category of things, so you could
infer that they are used for such things, such as ISBN numbers. "Ode to
Joy" qualifies as a title, a natural language noun phrase. Therefore we
deliberately did not regard it as instance of E75. You may however
argue, that "Ode" is indeed a category of E28. I have the impression
that we cannot bind form with use in this way, and rather leave the
distinction to a
less committed "P2 has type".
On 6/9/2015 5:05 μμ, Franco Niccolucci wrote:
> Hi all,
> Among the examples of E28 Conceptual Objects there is (rightfully) Beethoven’s “Ode to Joy”. I would believe that “Ode to Joy” is the appellation of this object, as is also "“Ode an die Freude”” etc.
> However, the examples of E75 Conceptual Object Appellation (related to E28 by P149 is identified by) include as examples only identifiers, i.e. instances of E42 Identifier, which is a specific kind (actually a subclass) of E41 Appellation. The same happens, more or less, in the examples of P149.
> My question is:
> - is this deliberate, and only identifiers are allowed for E75, as in the examples
> - or, “Ode to Joy” would also be an acceptable instance of E75, as results from the scope note?
> I will assume the latter. If I am right, adding “Ode to Joy” as an example of E75, i.e. an appellation that is not an identifier, would clarify.
> By the way, and in general, I am not much comfortable with all the story of Identifiers/Appellations.
> For example, in the scope note of E44 Place Appellation the term “identifier” is loosely used to define the class instances, but this may be misleading as appellations like “next door on the left after Franco's house, just before the Indian Restaurant named Kashmir” are not at all an identifier, but indeed a valid place appellation. In Japan this one may reportedly be used also as E45 Address, again called an “identifier” in its scope note. An identifier is an appellation, but an appellation is not (always) an identifier.
You mean: "This class comprises any sort of identifier
characteristically used to refer to an E53 Place." This must be changed!
This is an issue. So, I propose Place Appellation should be reviewed.
The "Section Definition" is also a constant source of confusion.
The idea is that an identifier must not be a natural language form or a
proper name for persons. The Question, to which degree a path
instruction is a place name, is indeed an issue to be discussed!
A Japanese address a I know it, is a hierarchical sequence of fixed
levels of placenames that goes down to the "chome" level, if I am right,
the block of houses surrounded by roads. The house numbers go around the
block, and not the road. This is why Japanese maps do not need road indices.
> Perhaps more important, there is an activity of Identifier Assignment (E15) as subclass of Attribute Assignment (E13), but there is no similar class of Appellation Assignment, which would include giving names, addresses etc. So one has to go up one level and use E13, possibly with a type specification, to describe the activity of associating names with things, which is still more frequent than assigning a DOI.
> While Identifiers are indeed important because, among others, they may enable Linked Data, I don't really care of knowing how the identifier was assigned, so a shortcut property would probably suffice in most cases to link the thing to its identifier, preferred or not. On the other hand, I’d often like to know by whom/when/why a name (an appellation) was given to something, and the full path including the activity would allow documenting who did it (participated in the activity in the role of author), why (was motivated by), etc. It is unclear to me why Identifiers have the privilege of a specific assignment activity, while poor Appellations don’t.
Well, there are people who do care how the identifier was assigned. This
is one of the old guide lines of the CIDOC Information Categories. It is
vital to trace objects in inventories and provenance of museum objects,
sometimes of immense value, and a key element of librarianship , whereas
the general assignment of a name is not subject of an important
historical record, as the use is not binding for a particular community.
Identifiers, by virtue of being intended to be unique, fulfill a
particular function in our society.
Since the CRM is strictly evidence based on existing data structures,
such arguments of symmetry , i.e., where is the appellation assignment,
do in general not hold. We have not found databases recording name
assigments in general. However, we do have databases recording name use.
Therefore, FRBRoo, modelling FRAD, introducedF52 Name Use Activity.
If you can provide more evidence of recording naming events, you can
propose a new class. Otherwise, all CRM users are free to extend:-) .
Thank you very much for your comments!
> Thanks for the attention
> Prof Franco Niccolucci
> Director, VAST-LAB
> PIN - U. of Florence
> Scientific Coordinator
> ARIADNE - PARTHENOS
> Piazza Ciardi 25
> 59100 Prato, Italy
> Crm-sig mailing list
> Crm-sig at ics.forth.gr
Dr. Martin Doerr | Vox:+30(2810)391625 |
Research Director | Fax:+30(2810)391638 |
| Email: martin at ics.forth.gr |
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 |
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Crm-sig