[Crm-sig] A symbol made of symbols

Martin Doerr martin at ics.forth.gr
Fri Jan 17 16:01:31 EET 2020


Dear All,

As a general remark, proposing different semantics for avoiding punning 
should never be done. It would put syntax over meaning, and that is the 
hell of semantic incompatibility since the invention of databases, and 
the reason why formal ontologies were invented. Multiple instantiation 
is an official feature of RDF, and the only means to reduce the 
declaration of all class combinations up-front, impossible in an 
open-world, and without telling the world something new. Only semantic 
clarity warrants a long validity of an ontology and its applications.

So, the question should be, is it "composed of" or "incorporates". I 
basically follow the reasoning George provided. It is much up to the point.

The actual discussion, if "Mark" is the "ur-image" or not, took place in 
Washington for the first versions of the CRM. It is definitely the 
"ur-image". The argument was that museums have identifiers in reference 
documents for them, and want to link directly. At that time, providing a 
digital image as intermediate was still exotic. Also, experts talking 
about the "same" mark, in contrast to an "identical" mark, would refer 
to the ur-image and its identifier.

We should add that to the scope note of E37. We thought the examples 
would render the distinction:-[

We may be relaxed about the symbolic composition of a monogram, and 
regard it to be composed of other symbols. (Letters *ARE* E90 Symbolic 
Object, not linguistic objects).

We should be more precise, and regard that the letters "P X" are 
"incorporated", because the graphical ideal is not a sum of its parts, 
but a particular image.

In George's graph, the chain should be: Physical Feature - is 
represented by Image - incorporates Mark - (is represented by SVG...), 
Mark is identified by "Reference Code"...

If there are Mark variants, they would still be "ur-images" of each 
variant, still Marks.

Opinions?

Best,


martin


On 1/16/2020 10:21 PM, Ethan Gruber wrote:
> Hi Rob,
>
> I'm not sure that works since we've decided that a monogram is an 
> E37_Mark ("This class comprises symbols, signs, signatures or short 
> texts applied to instances of E24 Physical Man-Made Thing by arbitrary 
> techniques in order to indicate the creator, owner, dedications, 
> purpose, etc.").
>
> I think your example doesn't allow us to answer the baseline research 
> question of querying for individual letters that comprise a monogram.
>
> <chi-rho> a crm:E37_Mark ;
>   crm:P106_is_composed_of "Χ"; #Greek chi
>   crm:P106_is_composed_of "Ρ" . #Greek rho
>
> Where is P190 documented? I'm looking at the PDF for 6.2.3, and I'm 
> not seeing that property in there. Or, there is a P190, but it's not 
> has_symbolic_content.
>
> Ethan
>
> On Thu, Jan 16, 2020 at 2:07 PM Robert Sanderson <RSanderson at getty.edu 
> <mailto:RSanderson at getty.edu>> wrote:
>
>     Ethan,
>
>     Could you do :
>
>     ?monogram a E33_Linguistic_Object ; crm:P106_is_composed_of
>     ?character .
>
>     ?character a E33_Linguistic_Object ; crm:P190_has_symbolic_content
>     “☧” .
>
>     ?
>
>     That would avoid the punning that the chi-rho is both an E33 and a
>     literal at the same time.
>
>     Rob
>
>     *From: *Crm-sig <crm-sig-bounces at ics.forth.gr
>     <mailto:crm-sig-bounces at ics.forth.gr>> on behalf of Ethan Gruber
>     <ewg4xuva at gmail.com <mailto:ewg4xuva at gmail.com>>
>     *Date: *Thursday, January 16, 2020 at 9:59 AM
>     *Cc: *"crm-sig at ics.forth.gr <mailto:crm-sig at ics.forth.gr>"
>     <crm-sig at ics.forth.gr <mailto:crm-sig at ics.forth.gr>>
>     *Subject: *Re: [Crm-sig] A symbol made of symbols
>
>     I have a followup question to the use of
>     crm:P165i_is_incorporated_in. We have implemented this property to
>     link a Monogram to a representative, idealized SVG URI. In a very
>     narrow subset of cases (maybe only one that I know of so far), a
>     monogram is notable enough to have warranted entry into Unicode,
>     the chi-rho Christogram: ☧
>
>     We have a need to define URIs for these Christograms so that we
>     can exploit the constituent letters via P106_is_composed_of in
>     SPARQL. We have at least a few examples of Monograms that consist
>     of both Latin letters and a Christogram, e.g.,
>
>     ?monogram crm:P106_is_composed_of+ "Ρ" #Greek rho
>
>     So I just want to confirm that a single Unicode character itself
>     is an E73 Information Object, even if this is an unusual
>     implementation.
>
>     ?monogram crm:P165i_is_incorporated_in "☧"
>
>     Ethan
>
>     On Fri, Nov 8, 2019 at 10:15 AM Ethan Gruber <ewg4xuva at gmail.com
>     <mailto:ewg4xuva at gmail.com>> wrote:
>
>         Hi George,
>
>         I think this makes a lot of sense. I can use the D1 Digital
>         Object, and this is pretty useful for us as I would like to be
>         able to associate the SVG with the person who created it or
>         other processes of production (derived from a font file,
>         e.g.). I've forwarded to the Nomisma list and hopefully we'll
>         agree and start publishing our monograms online soon.
>
>         Thanks,
>
>         Ethan
>
>         On Fri, Nov 8, 2019 at 6:28 AM George Bruseker
>         <george.bruseker at gmail.com <mailto:george.bruseker at gmail.com>>
>         wrote:
>
>             Hi Ethan,
>
>             Here is my take.
>
>                 I have a large number (thousands) of monograms that
>                 appear on Greek coinage. There is an SVG file that
>                 represents an idealized form of the monogram. The
>                 Nomisma ontology has a nmo:Monogram class, and I am
>                 attempting to link Nomisma more directly as subclasses
>                 or subproperties to CIDOC-CRM ones. A monogram fits
>                 the definition of a subclass of crm:E37_Mark:
>
>                 "This  class  comprises  symbols,  signs,  signatures
>                  or  short  texts  applied  to  instances  of  E24
>                  Physical Man-Made Thing by arbitrary techniques in
>                 order to indicate the creator, owner, dedications,
>                 purpose, etc."
>
>             Yes, it seems the right match.
>
>
>
>                 In this sense, if I want to link a monogram to its
>                 constituent letters, is P106_is_composed_of the
>                 appropriate property for this?
>
>                 For example, I have a URI for a monogram,
>                 http://numismatics.org/ocre/symbol/monogram.ric.10.theodosius_ii.3
>
>                 Therefore:
>
>                 <http://numismatics.org/ocre/symbol/monogram.ric.10.theodosius_ii.3>
>                 a nmo:Monogram ;
>
>                 crm:P106_is_composed_of "T" ;
>
>                 crm:P106_is_composed_of "H" .
>
>             This also seems the right match. If you are not concerned
>             about the particular form of the letters, then I guess you
>             could make the letters instances of E90 Symbolic Object.
>
>
>
>                 etc.
>
>                 The next question I have is how do I link this concept
>                 of a monogram to one or more SVG files that represent
>                 this monogram? There could be variant images based on
>                 individual styles of die-carvers, but scholars agree
>                 these variations represent the same semantic concept.
>
>                 I am looking at the documentation for P138 represents,
>                 and I am having a difficult time understanding the
>                 distinction between the examples where a digital file
>                 (PLY 3D model or a JPEG image) is the E36 Visual Item,
>                 but in other documentation the E36 Visual Item seems
>                 more conceptual.
>
>                 If a Visual Item is definitionally an E1 CRM Entity,
>                 then a Visual Item can still represent another Visual
>                 Item, correct? So:
>
>                 <http://numismatics.org/ocre/symbol/monogram.ric.10.theodosius_ii.3>
>                 a nmo:Monogram ;
>
>                 crm:P106_is_composed_of "T" ;
>
>                 crm:P106_is_composed_of "H" ;
>
>                 crm:P138i_has_representation
>                 <http://numismatics.org/ocre/symbols/monogram.ric.10.theodosius_ii.3>
>                 #svg file url
>
>             For the question of relating the instance of Mark (the
>             monograms) to the SVG, I would do this otherwise. I would
>             take advantage of D1 Digital Object class for the
>             instances of SVG and their characteristics. [if you won’t
>             like extensions, then E73 information object] I would then
>             link the instances of D1 to the individual marks through
>             the p165
>             <http://www.cidoc-crm.org/Property/P165-incorporates/Version-6.2.1> incorporation
>             property which allows one information object to
>             incorporate another.
>
>             For the question of relating one instance of Mark (such
>             that that is uniquely identifiable from another but which
>             is nevertheless a variant of the same Mark), you could
>             make use of the p130
>             <http://www.cidoc-crm.org/Property/P130-shows-features-of/Version-6.2.1> property
>             ’shows features of’. It has a property on property that
>             allows you to specify the kind of similarity.
>
>             I attach an example of the proposed solution as a diagram.
>             I guess the one part of your problem that it does not
>             address is the ur-imageness of the one idealization. I
>             guess the ur type did not historically exist but is the
>             composite based on scholarly research. Therefore it sounds
>             like creation of a type, see E83
>             <http://www.cidoc-crm.org/Entity/E83-Type-Creation/Version-6.2.1>
>             Perhaps this is a picture for a type? Or you could make
>             one instance of Mark which is the ur instance and say that
>             all the other instance are related to it in particular as
>             variant, but that doesn’t seem correct at first thought.
>
>             Best,
>
>             George
>
>             cid:16e4b932ae91cc3c9f51
>
>
>
>                 Thanks,
>
>                 Ethan
>
>                 _______________________________________________
>                 Crm-sig mailing list
>                 Crm-sig at ics.forth.gr <mailto:Crm-sig at ics.forth.gr>
>                 http://lists.ics.forth.gr/mailman/listinfo/crm-sig
>
>     	
>
>     *CAUTION: This email originated from outside of the Getty. Do not
>     click links or open attachments unless you verify the sender and
>     know the content is safe.*
>
>
>
>
> _______________________________________________
> Crm-sig mailing list
> Crm-sig at ics.forth.gr
> http://lists.ics.forth.gr/mailman/listinfo/crm-sig


-- 
------------------------------------
  Dr. Martin Doerr
               
  Honorary Head of the
  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
  
  Vox:+30(2810)391625
  Email: martin at ics.forth.gr
  Web-site: http://www.ics.forth.gr/isl

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ics.forth.gr/pipermail/crm-sig/attachments/20200117/f309c559/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 73609 bytes
Desc: not available
URL: <http://lists.ics.forth.gr/pipermail/crm-sig/attachments/20200117/f309c559/attachment-0001.png>


More information about the Crm-sig mailing list