Issue 276: FOL representation

Starting Date: 
2015-02-12
Working Group: 
3
Status: 
Open
Background: 

In the 32nd joined meeting of the CIDOC CRM SIG and ISO/TC46/SC4/WG9 and the 25th FRBR - CIDOC CRM Harmonization meeting, closing the issue 257, the crm-sig assigned to Christian Emil to prepare a list of shortcuts in order to be expressed in FOL.

Oxford, February 2015

Current Proposal: 

Posted by Christian Emil on 4/5/2015

In June 2014 Martin asked for volunteers to scan through the definition of CIDOC CRM and look for so called strong shortcuts. A strong shortcut  is defined as a shortcut from which we can imply the existence of the node(s) in the full path.

This is not a formal issue but I send the result of my small exercise to the list for discussion and comment.  The "strongness", that is, the implied existence of intermediate nodes depends in some cases on a given interpretation  of the model.

Not strong is called weak in the list below.

Kind regards,
Christian-Emil
************************************************************
Introduction

In this exercise I searched the scopenotes in the CRM 6.0 definition for the string 'shortcut'.  In several cases the word was used in the definition of properties that can be a part of the expansion of a short cut. For example, the "P2 Has Type" property is not defined as a short cut, but is referred to from other  scopenotes as a shortcut.

*******************************************************
GROUPS OF SHORTCUTS
*******************************************************

GROUP 1, observations/statements

The three shortcuts  have long paths through  sub classes of E7 Activity. Whether  the existence of such an intermediate node is implied by the existence of (an instance of) the shortcut depends on the actual use of the model and the users ontological commitment. If the type of a person (gender for example) is a permanent trait that need not to be assigned explicitly then there will be no full path, similar for condition state, dimension and identifier.  If ontological commitment implies that the assignment of types, condition state, dimension and identifier always are  results of (human) observation  and actions  like deduction, the shortcuts may be considered strong.

P2 has type (is type of)
Domain:  E1 CRM Entity
Range:   E55 Type
Shortcut through  E17 type assignment.

P43 has dimension (is dimension of)
Domain: E70 Thing
Range:  E54 Dimension
Shortcut through: E16 Measurement

P48 has preferred identifier (is preferred identifier of)
Domain: E1 CRM Entity
Range:  E42 Identifier
Shortcut through:  E15 Identifier Assignment

P44 has condition (is condition of)
Domain: E18 Physical Thing
Range:  E3 Condition State
Shortcut through: E14 Condition Assessment
*********************************************************
GROUP 2 
These are shortcuts with a long path through activities  that  change state in a wide sense (ownership, custody, place, membership).

**********
A

For an instance of E10 Transfer of Custody  the from and to (E39 Actor) are optional,  cardinality constraints (0,1:0,1)  [which in fact seems a little strange. What is the purpose of documenting a change of custody  when the to and from (E39 Actors) are undefined] . In the case of P49 and P50  one knows the receiver. The existence of the intermediate node is  implied. It is not unique. The properties P49 and P50  are strong shortcuts.

P49 has former or current keeper (is former or current keeper of)
Domain: E18 Physical Thing
Range:  E39 Actor
Shortcut through: E10 Transfer of Custody

P50 has current keeper (is current keeper of)
Domain: E18 Physical Thing
Range:  E39 Actor
Shortcut through: E10 Transfer of Custody

**********
B

For an instance of E8 Acquisition the from and to (E39 Actor) are optional, cardinality constraints (0,1:0,1)   [which in fact seems a little strange. What is the purpose of documenting a change of ownership when the to and from (E39 Actors) are undefined. In the case of P51 and P52 one knows the receiver. The existence of the intermediate node is  implied. It need not to be unique. The properties P53 and P55  are strong shortcuts.

P51 has former or current owner (is former or current owner of)
Domain: E18 Physical Thing
Range:  E39 Actor
Shortcut through: E8 Acquisition
P52 has current owner (is current owner of)
Domain: E18 Physical Thing
Range:  E39 Actor
Shortcut through: E8 Acquisition

**********
C

For an instance of E9 Move the from and to (P27 moved from, P26 moved to E53 Place) are necessary with the cardinality constraints (1,n:0,n). The existence of the intermediate node cannot be implied since the  Physical Thing may have been at the same place since it come into existence . The properties P53 and P55  are weak shortcuts.

P53 has former or current location (is former or current location of)
Domain: E18 Physical Thing
Range:  E53 Place
Shortcut through: E9 Move

P55 has current location (currently holds)
Domain: E19 Physical Object
Range:  E53 Place
Shortcut through: E9 Move

***********
D
A group is not created with members. Therefore there must be at least one event E85 Joining. The P107 shortcut is strong.
P107 has current or former member (is current or former member of)
Domain: E74 Group
Range:  E39 Actor
Shortcut through: E85 Joining

********************************************************
GROUP 3

The full path  for the shortcut below is:
E19 Physical Object through P59 has section (is located on or within), E53 Place, P53 has former or current location (is former or current location of) to E26 Physical Feature.  The instance of E26 Physical Feature must have  a location (place) which is located on  the instance of E19 Physical Object. Therefore the shortcut implies the existence of the intermediate node (E53) Place.  P56 is a strong shortcut.

P56 bears feature (is found on)
Domain: E19 Physical Object
Range:  E26 Physical Feature
Shortcut through: E53 Place
*******************************************************
GROUP 4
It is possible to state that a place is located on or within a physical object without an explicit E46 Section Definition (E41  Appellation).  P59 is a weak shortcut.

P59 has section (is located on or within)
Domain: E18 Physical Thing
Range:  E53 Place
Shortcut through: E46 Section Definition

********************************************************
GROUP 5
The definition of depict in OED: "To draw, figure, or represent in colours; to paint; also, in wider sense, to portray, delineate, figure anyhow." It depends on the ontological commitment which could be more explicitly stated in the scope note of P62 depict, something like: In CIDOC-CRM the existence of an information object/ E36 Visual Item representing the depiction is assumed/implied. In this case the existence of at least one  intermediate node is implied. The intermediate node is not unique. P62 is strong under the given ontological commitment. The scopenote should be made more specific about this.

P62 depicts (is depicted by)
Domain: E24 Physical Man-Made Thing
Range:  E1 CRM Entity
Shortcut through: E36 Visual Item

********************************************************
GROUP 6
The name of the P105 is somewhat misleading. It states  _a_ right is held by  (has _a_ right). This right must exists and the existence of an  intermediate node is implied. The P105 is a strong shortcut.

P105 right held by (has right on)
Domain:  E72 Legal Object
Range:  E39 Actor
Shortcut through: E30 Right

********************************************************
GROUP 7
This property may be a shortcut for two different paths (mother father)  through the birth event. This is not the case for non-biological parenthood. Thus it is not strong. P152 is a weak shortcut.
P152 has parent (is parent of)
Domain: E21 Person
Range:  E21 Person

 


Martin and Christian Emil  have sent on 8/5/2015 the following list of shortcuts in order to be discussed in the forthcoming meeting in Nuremberg at  19/5/2015


In 33rd CIDOC CRM meeting in Nuremberg, the crm-sig discussed about shortcuts formalization which had two action points (1) to control all the quantifications and (2) all subproperties , they made comments to the proposed list. The comments are in the minutes of this meeting.

The outcome was:

P2 :This is a weak shortcut.

Action: We have to distinguish knowledge. Øyvind will make a statement about knowledge creation process and shortcuts. If one who maintains the knowledge base takes a stamp of the observer. Knowledge creation events cannot be inferred by rule. (if you are the God, you don’t describe yourself in the graph you may describe other gods). George Bruseker  and Øyvind will work in this. 
P43 : It is a weak shortcut, dimension could be computed
P44: This shortcut concerns the epistemological layer
P48:  This is not a shortcut. There is a cardinality conflict!: It is decided that Martin and CEO will check the FOL for shortcut Logic
P49:  A comment should be stated. Steve, MD, Athinak should think together
P50: it is not a weak shortcut
P51: It is a weak shortcut, 

P52: it is not a weak shortcut.

P53: Action : to be discussed in the sense of causality. It is complicated  because of E53 Place. The inverse is weak because create the shortcut from the link and not from the whole path. It should be discussed.

P55: It is a weak shortcut, alternatives are ambiguous

P56: It is a strong shortcut
P59: It is a weak shortcut, alternatives are ambiguous
P62: to be elaborated by Steve

P105: There is a problem with the “current” when data is aggregated. There is no problem with local data.  Maybe an addition is needed to CRMinf to propose how a property is current. Wolfgang Schmidle and Pierre Choppe, will send what they have understood. 

P107: It is a weak shortcut.  The sig reviewed the scope note of E66 Formation and it is noted that this doesn’t mean that the members of the group are being enumerated.  It is highlighted a part of the scope note, in order to be discussed later. The question to Carlo is how to formulate that an inverse exist in strong shortcuts? Then the sig made changes to the scope note of E85 Joining.
P152: It is a weak shortcut. Alternatives are ambiguous. The question is what are the states of not knowing? We have to revise what is the utility of making a strong shortcut inference explicit (question to Carlo)
P130: MD will change the scope note to reflect the inference
P109: The scope note is changed. “ The sig decided that  in the introduction to CRM it should be stated that someone  before read and use CRM, should read certain documents. It should be written an introduction for RDF and OWL representation of CRM. (Mark Fichtner will prepare such documents)
P8:  The scope note is changed. It is a weak shortcut
 

Nuremberg, May 2015

 

In 34th crm-sig meeting, the outcome of the discussion about shortcuts are

P2: This is a Weak shortcut! ,The proposed text about knowledge creation process has been accepted.
P48: This is not a shortcut. There is Cardinality conflict!   Martin and CEO will check the FOL for shortcut Logic
P49: The scope note is changed as well as the scope note of E10
P53 :It should be discussed. 
P62: to be elaborated by Steve

P105: There is a problem with the “current” when data is aggregated. There is no problem with local data.  Maybe an addition is needed to CRMinf to propose how a property is current. Wolfgang Schmidle and Pierre Choppe, will send what they have understood. 

P107: It is pending the   question to Carlo " how to formulate that an inverse exist in strong shortcuts? " 
P152:It is pending  he   question to Carlo  "what are the states of not knowing? We have to revise what is the utility of making a strong shortcut inference explicit "
P130: It is a conditional shortcut. The scope note is revised. The inverse subproperty should be revised.
P109: It is pending " The sig decided that  in the introduction to CRM it should be stated that someone  before read and use CRM, should read certain documents. It should be written an introduction for RDF and OWL representation of CRM. (Mark Fichtner will prepare such documents)"

Also we decided that we need a statement about “What  a knowledge base is.”, this text will be elaboratied by Carlo. This is will be a new issue (see issue 297)

Finally, we decided that we need an explanatory text for FOL representation to be added in the terminology section of CIDOC CRM.

Heraklion, October 2015

In the 35th joined meeting of the CIDOC CRM SIG and   28th FRBR - CIDOC CRM Harmonization meeting the crm-sig accepted the headers for OWL version of CIDOC CRM (see issue 281). The proposed headers  are described in this file


Posted by Steve in 22/7/2016

Revision of Scope note:

BEFORE:-

P62 depicts (is depicted by)

Domain:                E24 Physical Man-Made Thing

Range:                  E1 CRM Entity

Quantification:  many to many (0,n:0,n)

Scope note:        This property identifies something that is depicted by an instance of E24 Physical Man-Made Thing. Depicting is meant in the sense that the surface of the E24 Physical Man-Made Thing shows, through its passive optical qualities or form, a representation of the entity depicted. It does not pertain to inscriptions or any other information encoding.

This property is a shortcut of the more fully developed path from E24 Physical Man-Made Thing through P65 shows visual item (is shown by), E36 Visual Item, P138 represents (has representation) to E1CRM Entity. P62.1 mode of depiction allows the nature of the depiction to be refined.

Examples:         

§  The painting “La Liberté guidant le peuple” by Eugène Delacroix (E84) depicts the French “July Revolution” of 1830 (E7)

§  the 20 pence coin held by the Department of Coins and Medals of the British Museum under registration number 2006,1101.126 (E24) depicts Queen Elizabeth II (E21) mode of depiction Profile (E55)

AFTER:-


P62 depicts (is depicted by)

Domain:                               E24 Physical Man-Made Thing

Range:                  E1 CRM Entity

Quantification:  many to many (0,n:0,n)

Scope note:        This property identifies something that is depicted by an instance of E24 Physical Man-Made Thing. Depicting is meant in the sense that the surface of the E24 Physical Man-Made Thing shows, through its passive optical qualities or form, a representation of the entity depicted. "Passive optical qualities" specifically excludes anything that actively transmits light. It does not pertain to inscriptions or any other information encoding.

This property is a shortcut of the more fully developed path from E24 Physical Man-Made Thing through P65 shows visual item (is shown by), E36 Visual Item, P138 represents (has representation) to E1CRM Entity. P62.1 mode of depiction allows the nature of the depiction to be refined.


posted by Franco on 22/7/2016

I recently reviewed a (bad) paper on multimedia which revamped my interest on the topic. Let me tell you my doubts-

P62 concerns an E24 Physical Man-Made Things, together with P65 shows visual item, although the latter is well known to have a slightly different meaning. E.g.: a caricature on paper, Mona Lisa, a coin with Queen Elizabeth’s profile, a postcard of Crete, the photo of Martin Doerr when he was 6 years old (yes, Martin has been a child, but long time ago): they are all allowed to P62 depict. So far, so good.

If today I take a digital photo of Martin Doerr, e.g. with my iphone, what I produce is an E73 Information Object, stored somewhere in the iphone, in the memory of my Mac where I copy it, in the iCloud where I regularly back up the iphone content etc. This E73 exists as long as at least one of these physical carriers (E84 Information Carrier) stores it. E84 is an E22 Man-Made Object, a subclass of E24, so it is an E24 as well. As such, it is allowed to P62 depict.

Going up the E73 class genealogy, (E73 subclass of E89 subclass of E28) one arrives at E28 Symbolic Object, so the file produced on my iphone is an E28, which is the sister of E24, and both are distinct siblings of E71 Man-Made Thing. Being immaterial, E73 is NOT allowed to depict.

In conclusion: if now I take a photo of Martin with a film, the film (and any printout out of it) P62 depicts Martin. But if I take the same photo with my iphone, what depicts Martin? Not the corresponding jpeg file E73 in the iphone (or any copy of it), because E73 does not belong to the domain of P62. Possibly the iphone memory where the file is stored, which as physical thing inherits the P62 domain from E24, but it sounds a bit weird to say “Franco’s iphone memory card depicts Martin Doerr".

Any solution to this asymmetry?

Note that both things, the film and the file, are related to the same visual item “image of Martin Doerr on 22/07/2016”: but the film is allowed to P65 show it, the file is not.

Why not defining the domain of P62 as E71 Man-Made Thing, to incorporate both material items, coinciding with their carrier, and immaterial items, stored on a separate carrier? when we look at a painting, do we look at the assemblage of canvas, pigments, etc or we look at the E36 Visual Item P65 shown on it? They seem to be inseparable, but as digital technology shows, possibly they are not.

This may lead to the question: what about icloud, is it a physical thing or what? It is of course made of physical disks, flash memories, cables etc, but without the appropriate software all this is just a heap of scrap iron that can’t store anything.
But let us keep this question for another thread.

Thanks in advance for any comment on the above concerns, they keep me awake in the nights of this hot July.


posted by Christian Emil on 23/7/2016

A file is nota E73 information object. It is a physical object. You may store your portrait of Martin on a punch tape (of a considerable length) or a stack of punch cards for that matter. In principle there is no difference between an object with a magnetic emulsion and a punch tape with respect to carrying information.

Yu ma also go back to the polyphone (http://www.hlxx.de/hp/polyphon.htm) which  indeed by many have been considered to be a heap of scrap metal and pieces of wood. Still it is able to store information.


posted by Daria Hookk on 23/7/2016

I know that we have metadata about original image inside of a file with the prtrait. So, they mark which copy is original.
If I change size or other changes - metadata will be desroyed. Thus, image made by my Nikon S1 or by iphone -original - all others - copies, because they are a result of my activity titled copying.


posted by Franco on 23/7/2016

Thank you Christian-Emile for your comment and explanation.

If you are right, then the scope note of E73 must be amended:

"This class comprises identifiable immaterial items, such as a poems, jokes, data sets, images, texts, multimedia objects, procedural prescriptions, computer program code, algorithm or mathematical formulae, that have an objectively recognizable structure and are documented as single units.
[. . .]
Examples:
-  image BM000038850.JPG from the Clayton Herbarium in London
[. . .]”

(quoted from page 28, cidoc_crm_version_6.0-2.doc)

I understood that this BM000038850.JPG is a file, otherwise what is it?

In my opinion - but I may be wrong - your punched tape, or cards, or my hard disk, are instances of E84 Information Carrier; the file is what is written on them:
E84 Information Carrier “This 20-km-long C-E Ore’s punched tape” P128 carries E73 Information Object “photo12345.jpg, a pretty and rare picture of Franco Niccolucci in shorts”.

Another example: given its rarity, I now expect to receive messages from all the SIG members stating “Please send me by email the file of your photo in shorts”. I do not think they will mean to receive the punched tape, a bit too heavy to attach to an email; rather (its content as) an immaterial file attachment.

The confusion may arise from the linguistic (ab)use of the term “file”, which is, according to my Oxford dictionary:

1) a folder or box for holding loose papers that are typically arranged in a particular order for easy reference
2) the contents of a file folder or box
3) a collection of data, programs, etc., stored in a computer's memory or on a storage device under a single identifying name

1) and 2) are likely to be E22 Man-Made Object and, since we keep it in order (well, not always in good order) for the information it contains, E84 Information Carrier
3) is more likely to be an E73

This is confusing in English: in Italian (and Dutch) a “file” is only a computer file, with other words for 1)-2); in German, precise as usual, a computer file is a “Datei”, otherwise it is an “Ordner"; anglophobic Spanish and French have adopted old terms to new technology; etc. Maybe people’s thinking is influenced by their native language usage, and in all these languages “file” sounds only as immaterial computer stuff, data and the like.


posted by Christian-Emil on 23/7/2016

The punch tape, polyphone disk and  vinyl record are all very concrete examples of the difference between the carrier and the abstract content. A magnetic tape be it for computers, music or video (e.g VHS) may also be carriers of information although it gets more abstract since we cannot see the patterns. On a magnetic disk or a flash drive the patterns representing/carrying the information is even more hidden.  In all cases it is the physical pattern on the physical object that carries the information.

In the (early) 1980s "“Please send me by email the file of your photo" would not have been relevant. However you could get a copy of  the file on a magnetic tape (or a copy of the original punch tape) or using ftp. In all cases there will be a process of copying the pattern from one medium to another and by this changing the receiving medium in the same way a laser printer change the paper by adding toner to it. The paper  with the toner pattern  is a physical object. The email + the file will  receive your computer through a series of copying processes.

I think you are quite right when pointing to  everyday  languages. We need to use care when we formulate the scope notes. Remember also that the names of the classes in CRM are just there as a crutch for the memory. The meaning of a class is in the scope note solely.


posted vy Steve on 23/7/2016

I think that everyone has got a little off track here!
E73 Information Object is immaterial.
The instance of E22 Man Made Object {Franco's iPhone} P128 carries (is carried by) the instance of E36 Visual Item (a sub-class of E73 Information Object) {Photo of Martin in shorts} P138 represents (has representation) E21 Person {Martin}..
Should Franco, as a mark of respect, have the image laser etched into the back of his iPhone then:-
E22 Man Made Object {Franco's iPhone} P62 depicts (is depicted by) E21 Person {Martin}.
All the other instances of carrier (Franco's Mac, punch tape, polyphone Disk, vinyl record, punch cards, hard drives and memory card) would use P128 to carry the digital file and P62 only if the image was etched or printed on the carrier.

The point of the edit of the P62 scope note was to ensure that no one thought that a device that projected the images was "depict"ing.

I think Franco might top-up his retirement fund if starts a range of memory cards with images of Martin depicted on them. I am sure they will sell well. Perhaps with he could start a range with the second special collector's edition using his own selfie!


posted by Steve on 23/7/2016

Answers in Line.

But the trick is to read the scope note not the natural language of the label with all it’s possible interpretations

HTH

SdS

 

From: Simon Spero [mailto:sesuncedu@gmail.com]
Sent: 23 July 2016 18:28
To: Stephen Stead <steads@paveprime.com>
Subject: Re: [Crm-sig] P62 Homework

Does the painting depict the revolution in the same sense that the coin depicts the queen? Yes

Does a portrait done using ink that is only visible using ultra-violet light depict? Yes

Does a ball-and-stick model of a double helix depict DNA? No

Does an e-ink image of a portrait of Charles Dickens on an e-reader depict the author? No

Does the same image displayed on an OLED e-reader display not depict the author? Correct it does not

If the display is left static for too long, and causes burn in, does the answer change? If the image can be seen with the device powered off then yes if it requires active optical qualities then no

If an image of a portrait of Harper Lee is shown on the screen with burn-in, does it only depict Charles Dickens? See above but if both images are visible then only Charles is depicted

Does an exhibit of a Rifle, Short, Magazine, Lee-Enfield Mk. III depict itself? No The class of all such rifles? No

What about an exhibit of such a rifle, but with a specific history (RCIN 69430,  used by T. E. Lawrence)? no

Do shadows of people created by the detonation of an atomic bomb depict? Yes

Does a stuffed Antelope in a museum depict, but a live Antelope not? Neither do
Does a live Antelope, in a museum, wearing a hat saying "trained mooseologist" , depict? Nothing If so, what? See above!


posted by Steve on 23/7/2016

If you make a change you generate a new E73 Information Object no matter how minor the change.


posted by Christian Emil on 24/7/2016

This an interesting issue that has links to visual arts. For example a series of lithographic prints (I know that each print is considered a unique piece of art) or more mundane positive copies of a photographic negative.

Also if I change the typeface in a word file, I get a new instance of E73 Information Object. On the other hand both the original and the new file on the disk may be considered to carry the information in one E73 Information object (the one where typeface is of no significance.

A more detailed discussion should in my view be added to the FAQ section of the definition of CRM.


posted by Martin on 24/7/2016

Dear Franco, All,

The property "depicts" was meant to do it via a visual process, in particular statues and paintings, that by their
whole shape and surface properties represent something. This means, by surface properties and passive light reflection. The electronic world was not in the primary scope of the CRM. If "depicts" should be used more broadly, is a question to be discussed.

I agree that a confusion comes from the term "file". It is simply, always wrong, to compare an ontology
with a word. The question is the identity condition. The E73 is something that is potentially IDENTICAL
on multiple carriers. A single carrier is a physical thing. A file content on a magnetic device is a magnetization
pattern, as such hardly visible. A CD could be read bit by bit by microscope. Both are not resembling in any way
what they are meant to represent. What we call "file" is not relevant for us, but which identity condition is meant for documentation.
A file content, the physical feature, is turned into visual signals by a complicated process, and this varies significantly from device to device, at least in color calibration and resolution. Therefore I'd prefer not to use the "depicts" property for encoded information.

Recently in the PARTHENOS Project, our team has discussed the concept of a "volatile dataset". We'll present
next week. It's identity condition is more that of a "Work", intentional rather than by encoding. 


posted by Simon Spero on 25/7/2016

On Sun, Jul 24, 2016, 3:39 PM martin <martin@ics.forth.gr> wrote:

    Dear Franco, All,

Dear Martin, Stephen, all :-)

    The property "depicts" was meant to do it via a visual process, in particular statues and paintings, that by their whole shape and surface properties represent something. This means, by
    surface properties and passive light reflection.


[all uses of "depict"  or "depicts" that follow should be understood as referring to P62, possibly with the second argument unspecified]

This roughly matches my understanding, which makes a couple of Stephen's answers confusing; possibly because my questions were unclear.

1)  a picture on an e-ink display does not depict.
This surprised me, as e-ink (and e-paper in general) work by passive light reflection, and only require power to change the display.
The intended contrast was with the active OLED display, which emits light, and requires continuous power.

2) a picture that requires a UV lamp to be seen does depict.
This question was aimed at clarifying whether the image must be produced by (subtractive) reflection of incident light, or if fluorescence caused absorption of that light was sufficient.

3) a ball-and-stick model of DNA is  not a depiction of DNA.

I am unsure why this is the case; it is a symbolic representation, created by human activity, and intended to be decoded using the human visual system without the assistance of specific equipment.
If it does not depict, then it is not clear that "Guernica" does. 

I assume it is uncontroversial that  "Photograph 51"  depicts DNA? 


posted by Steve on 25/7/2016

Simon

1] My bad I did not know the technology was passive.

2] I still think that this is passive.

3] This was me being late night picky! Yes the model depicts something but what? Is it the concept of DNA or a particular identifiable instance of DNA or …..

As to “Photograph 51” I am unfamiliar with this particular example of the photographers art but again, when you say DNA what do you mean?


Posted by Martin on 25/7/2016

Dear Simon,

On 25/7/2016 12:57 πμ, Simon Spero wrote:
>
> On Sun, Jul 24, 2016, 3:39 PM martin <martin@ics.forth.gr> wrote:
>
>     Dear Franco, All,
>
> Dear Martin, Stephen, all :-)
>
>     The property "depicts" was meant to do it via a visual process, in particular statues and paintings, that by their whole shape and surface properties represent something. This means, by
>     surface properties and passive light reflection.
>
>
> [all uses of "depict"  or "depicts" that follow should be understood as referring to P62, possibly with the second argument unspecified]
>
> This roughly matches my understanding, which makes a couple of Stephen's answers confusing; possibly because my questions were unclear.
>
> 1)  a picture on an e-ink display does not depict.
> This surprised me, as e-ink (and e-paper in general) work by passive light reflection, and only require power to change the display.
> The intended contrast was with the active OLED display, which emits light, and requires continuous power.

Well, the question is again not what means "depicts", but which defibnition is useful for cultural-historical reasoning. I'd argue that the thing on the screen of the e-paper is accidental to the device. So, it is simply inadequate to use a static property for what's at some instant on its screen. If it is passive or not, is not
the problem, but that is is not persistent to the object.
Further, an instruction how to draw something, in analogy to a file being instructions, cannot be regarded
a depiction in itself, I'd argue.

>
> 2) a picture that requires a UV lamp to be seen does depict.
> This question was aimed at clarifying whether the image must be produced by (subtractive) reflection of incident light, or if fluorescence caused absorption of that light was sufficient.
Well, why not, as long as it is an intrinsic property. Many visible colorants have fluorescence between visible
frequencies of light.

>
> 3) a ball-and-stick model of DNA is  not a depiction of DNA.
Well, here is a question of particulars and universals. I'd argue it does depict a structural abstraction of
some DNA molecule.

>
> I am unsure why this is the case; it is a symbolic representation, created by human activity, and intended to be decoded using the human visual system without the assistance of specific equipment.
> If it does not depict, then it is not clear that "Guernica" does. 
>
> I assume it is uncontroversial that  "Photograph 51"  depicts DNA?
What sort of photograph is this? Visible light would not reproduce molecular dimensions. If it could
it would not depict DNA, but some DNA molecule, a particular one, or not?


posted by Daria Hookk on 26/7/2016

Sorry, some ideas may useful or not.
"Depicts" must be permanent feature without electricity or other support, so paintins depicts, display depicts painting.
DNA by plastic balls - a model of an object like a copy of a cathedral.
File. In Russian it is also plastic bag for documents, but only in office context (please, give me one file to put the paper). Originally, of course, a set of bytes.


posted by Thanassis Velios on 1/8/2016

Another thought: Stephen's scope note works with stereograms as well:

https://en.wikipedia.org/wiki/Autostereogram#/media/File:Stereogram_Tut_...

Say this image was printed on a piece of paper. It passively depicts jumping horses and it passively depicts a shark. Both depictions are persistent, no problem. My only doubt is that the shark can only be seen/depicted after the viewer acts by following a viewing technique (i.e. crossing one's eyes) and without an instrument.

I think figurative kinetic art could be considered similarly: persistent depictions can only be observed while the viewer moves although I could not find a good example to share.

Maybe this is outside the scope of the CRM, but it is not obvious how "depict" is related (or whether it should be related) to the viewing event as opposed to being totally passive. 

Posted by Wolfang on 2/8/2016(current data)

This is what I could extract from my notes from the 2015 Nuremberg meeting.

 

The discussion about P105 was part of a discussion about shortcuts and "current" properties such as "P52 has current owner". Many shortcuts are "current" properties and at first glance seem to make statements about the present regardless of when the statement was actually added to the data. This poses a problem, especially if data is copied from one database to another or taken as a fact elsewhere as in the Semantic Web. In Cidoc 5.1 (2012) the "current" properties and their misleading labels were kept but the scope notes were amended to include "at the time of validity of the record or database containing the statement that uses this property".

 

P105 is also a "current" property. It's a shortcut and the longform contains P52 has current owner. However, it was overlooked because it doesn't have "current" in the label. So, the scope note needs to be amended, too (unless the scope note has changed since version 6.2.1 from October 2015).

 

Another point in the discussion were properties that are *not* current even though the present tense in the labels suggests otherwise, for example P43 has dimension and P44 has condition. The scope note of these two properties says "It offers no information about how and when an E54 Dimension / the E3 Condition State was established, nor by whom." There are more properties that have a present tense in the label but are presumably not current. An (incomplete) list where there is no comparable statement in the scope note:

 

* Physical Object P56 bears feature Physical Feature (a shortcut) and Physical Thing P59 has section Place (part of the longform)

* Physical Object P57 has number of parts Number (shortcut). The Physical Thing P46 is composed of Physical Thing (part of the longform) does contain "This property does not specify when and for how long a component element resided in the respective whole."

* Place P89 contains Place. However, I am not sure whether a place containing another place is a kind of "eternal fact".

 

It seems to me that the scope notes of these properties should be amended, too.

 

The only mention of CRMinf in my notes was: Use CRMinf to express "it was like this in 2014". Apparently .1 properties are not a good way of expressing currentness. Martin said that properties are never current, with a few exceptions for historical reasons. He argued that the notion of "current" should be an inferred information and not a property at all. As I understood it, there might still be a push to get rid of the "current" properties entirely if no good way of expressing currentness can be found. Steve suggested to deprecate them, which seems sensible to me.

In the 36th joined meeting of the CIDOC CRM SIG and ISO/TC46/SC4/WG9 and the 29th FRBR - CIDOC CRM Harmonization meeting, the SIG  discussed, resolved and made new homework assignments to the following items:

(a) P62: reviewed and close  the proposal made by Steve about the definition of P62.  

(b) Explanation about FOL representation: reviewed the text written by Carlo about the (FOL) logical expressions of the CRM and accepted it. This text should be pusblished   in a new section in applied form after the naming convention chapter.

(c) The sig accepted Steve’s proposal   saying that the definition of the property in the terminology section should be updated in order to indicate this aspect of timelessness with regards to properties. This HW is assigned to  Wolfgang, Steve and MD.

(d) P53 is a weak short cut and the weak shortcut axiom should be added and noted in this property. This HW is assigned to MD. Also it is proposed a systematic review of all the examples referenced to  places. This is assigned to  CEO, MD

(e) It is  proposed that there is a new issue examining the notions of  acquisition, ownership, ownership of the right  along with the notion of “current properties”

Iraklio, August 2016