chin-rcip / collections-model

Linked Open Data Development at the Canadian Heritage Information Network - Développement en données ouvertes et liées au Réseau canadien d'information sur le patrimoine
Creative Commons Zero v1.0 Universal
12 stars 1 forks source link

Fields and concepts to model for the Object Facet #71

Open stephenhart8 opened 3 years ago

stephenhart8 commented 3 years ago

As we are closing the 2.2 version of the TM, and focusing on the object facet for the next iteration, we need to establish the different fields and concepts that should be modelled in the TM. The goal of this issue is to identify those concepts, based on different data samples, and not to develop the patterns per se at this stage.

illip commented 3 years ago

Updated in order to reflect @TrangDg's comments and @marielmat's data:

  1. Identifier a. Manage identifiers from different provenances b. Shared identifiers for different components

  2. Collection a. Collection title b. Collection Description c. Collections in Collections d. Beginning and the end of the collection (could be computable)

Questions: a. What could be part of a collection? b. Should we focus only on physical things or intangible heritage are also considered as collection items? c. Should we document the related actors like a donor, a creator, etc.?

  1. Object Classification/Type

Questions: a. Merge those kinds of fields? b. How to manage the classification hierarchy(ies)?

  1. Object Appellation a. Title name b. Object name c. Generic object name d. Different titles depending on the context (exhibition, language, etc.)

Questions: a. What is the distinction between the Appellation and the Classification? b. Should we manage the generic object name (e.g. chair) has part of the classification?

  1. Exhibition

Questions: a. Should it be part of our current work?

  1. Subject a. Depicted subject(s) b. Categorization aspect (e.g. Art)

Questions: a. Should we document this in other more structured fields?

  1. Date and Period

Questions: a. Should we infer the Period using the date and the place? How to handle the different understanding of the period boundaries (e.g. Victorian Era or Meiji era)? b. Do we need to document the period's boundaries? c. Distinction between different beginning events (e.g. when the picture was taken, when the picture was developed, etc.) d. How to handle a label like "First quarter of the 18th century"?

  1. Related Actors a. Specific roles in different contexts (donors, creators, artists, etc.)

Questions: a. How to manage "in the style of"?

  1. Signature a. Physical marks description (color, etc.) b. Localisation c. Signature text

  2. Material, Medium and Support a. Information (artwork, etc.) material b. Support material (original and added by the curators)

Questions: a. Is there a clear distinction between both? b. Is there a clear distinction between medium and technique? (e.g. Albumen Process) c. Is it possible to distinguish them as they are often recorded in the same field with "Techniques"? d. How to link the materials and the techniques?

  1. Techniques a. used in the production to achieve the final product

  2. Place of production a. could also be place of origin of object

Questions: a. Would it be enough to reused only a place URI or sometimes it is useful to document the "place within a place" hierarchy (eg. Montreal, Quebec, Canada, North America) b. How to handle historical places?

  1. Description of object (physical) a. dimensions: width, height, depth, weight, size (when open/close) b. shape, color, scent, texture, etc. c. distinguish values from unit of measurement, e.g. 2 (value) metre (unit) d. condition: physical description or its own pattern?

  2. Description of object (textual)

  3. Supporting texts a. Notes/remarks/research/bibliography

  4. Partitioning of object a. an object can be composed from multiple components, e.g. a tea set includes a tea pot, cups and saucers b. each part itself can be described as a standalone object c. linking parts to whole and vice versa d. Number of components

Questions: a. Do we need to make a difference between a piece that you can remove from a part that you cannot?

  1. Location a. current display location and/or storage b. physical and/or digital (including object's URL?)

  2. Provenance a. type of transaction, date, place, parties involved (doner, buyer, seller, auction house, broker, etc.), monetary value (if applicable), identifier of transaction.

Questions: a. only its provenance prior to being acquired by data provider, e.g. donation by..., purchase from...; or an object's entire provenance if documented and available?

  1. Image a. Image of object b. URL

  2. Rights a. of images b. of object itself

  3. Style/Culture

  4. Publications a. Article where the item has been mentioned

    To be continued.... feel free to contribute and I will update this post accordingly! :)

TrangDg commented 3 years ago

Other fields besides @illip's list:

  1. Techniques a. used in the production to achieve the final product
  2. Place of production a. could also be place of origin of object
  3. Description of object (physical) a. dimensions: width, height, depth, weight, size b. shape, color, scent, texture, etc. c. distinguish values from unit of measurement, e.g. 2 (value) metre (unit) d. condition: physical description or its own pattern?
  4. Description of object (textual)
  5. Supporting texts a. Notes/remarks/research/bibliography
  6. Partitioning of object a. an object can be composed from multiple components, e.g. a tea set includes a tea pot, cups and saucers b. each part itself can be described as a standalone object c. linking parts to whole and vice versa
  7. Location a. current display location and/or storage b. physical and/or digital (including object's URL?)
  8. Provenance a. only its provenance prior to being acquired by data provider, e.g. donation by..., purchase from...; or an object's entire provenance if documented and available? b. type of transaction, date, place, parties involved (doner, buyer, seller, auction house, broker, etc.), monetary value (if applicable), identifier of transaction.
  9. Image a. Image of object b. URL
  10. Rights a. of images b. of object itself
  11. Style/Culture

Some comments on @illip's list:

  1. Collection: I think we need to clarify what is "Collection". a. not sure what you meant by "only physical things". d. yes, there would be case of collections within a larger collection e. yes, there would be case where a collection was donated by someone, created by the same actor, etc. f. Dates? (time-span derived from dates of objects in collection)
  2. Object Appellation
    • might need object names as not all objects has titles?
  3. Date and Period
    • it depends on what is considered to be Period? not hard if it's just 18th Century, but not for value like "Victorian era", "Meiji era", etc.
  4. Related Actors
    • specific to the production?
  5. Medium and Support
    • indeed it can be not clear as they are usually recorded together in the same field, even the tools (e.g. paper; pen and ink) and the technique (gelatin silver print). But if possible, the materials used to express the information layer of the works (ink, graphite, chalk, oil paint, acrylic, gelatin etc.) should be distinguished from the materials used for the support layer of the works (paper, canvas, silk, cardboard etc.). The support layer can have two types: primary support (usually part of the original works) and auxiliary/secondary support (usually added afterwards by conservators). Also, an object can be produced by different techniques and processes, so should the materials, and tools link to the techniques? However, it's understandable that such granularity is most likely not documented, and it's time consuming to clean them. :)
illip commented 2 years ago

Thank you @TrangDg, I tried to add all your ideas in my post to easily review the list with the Semantic Committee. I would also like to thank @marielmat and @Christian-McCord for their contribution.

I think we already have a great starting point in order to begin the Objects Facet. I didn't had the time to think that much about the overall categorization but perhaps we will have the time to discuss it tomorrow during the Semantic Committee meeting.

Speaking of the Objects Facet, CHIN would like to scope what we should be including in this facet. Perhaps the list below could help us to define the different types of content we would like to tackle and in which order. A few questions to nourish the reflection:

  1. Should we focus on the material vs. the intangible?
  2. Should we do the work from domain standpoints? (archeology, history, etc.)
  3. Should we go by objects types? (photography, artwork, etc.)
  4. Should we tackle everything at the same time? If not, how to prioritize?
  5. Should we start with a few common/transversal patterns?
marielmat commented 2 years ago

En vue de la 13e rencontre du comité sémantique, voici mes premières réflexions sur les points 1 à 10 résumés par @illip

1-2 a. What could be part of a collection?

Au MCQ, collection institutionnelle et collections rassemblées par certains donateurs. Mais l'interprétation est plutôt large, chacun ayant sa propre "définition". On peut parler de "la collection du musée", la "collection du Séminaire de Québe", la "collection de jouets", la "collection Brochu" etc. Nous avons des "secteurs de collectionnement" dont les définitions continuent d'évoluer avec le temps. Difficile d'arriver à un consensus. À mon avis, ce n'est pas sur ce terme que nous devrions commencer nos réflexions. Est-ce que les objets amassés ou collectés forment automatiquement partie d'une "collection"? Documenter le processus de collectionnement me paraît plus à propos que de rattacher un groupe d'items à une collection parce que x a dit que c'était une collection. Ce sont deux moments.

1-2- b. Should we focus only on physical things or intangible heritage are also considered as collection items?

Patrimoine culturel immatériel : https://ich.unesco.org/fr/qu-est-ce-que-le-patrimoine-culturel-immateriel-00003 Guide de documentation du réseau Info-Muse > champ "Type de patrimoine immatériel associé" > liste de types provient de l'IREPI (agricole, alimentaire, artisanal, artistique, autochtone, commercial, corporel, domestique, festif, forestier, industriel, institutionnel, ludique, maritime, mémoriel, militaire, musical, religieux, sportif, tradition orale, vestimentaire) = Qu'est-ce que ça veut dire?! Encore faut-il être capable de bien les définir, et vus de cette façon, les termes ne sont que des E55_Types documentés dans un E32.

À mon sens, le patrimoine immatériel doit être considéré dans les travaux, bien que ce patrimoine est documenté de façon à être conservé de manière "tangible" (voir le Musée de la mémoire vivante). Il y a beaucoup de récits associés aux objets. Ceux-ci sont parfois mal reflétés dans notre base de données, mais à mon avis ils font tout autant partie des collections et aident à les interpréter.

1-2 c. Should we document the related actors like a donor, a creator, etc.?

Pour ce qui est des donateurs et des donatrices, c'est confidentiel, on ne diffuse pas cette information. Le nom qui apparaît dans la zone de Mention n'est pas nécessairement celui des personnes qui ont donné les objets.

Pour le reste : oui. Ces personnes font partie intégrante du processus de production (le plus souvent) et de l'historique d'usage. Comme nous documentons généralement à partir de "l'objet", nous ne mettons pas nécessairement ces individus en relation et le fait de s'y attarder permettrait probablement de visualiser des liens insoupçonnés.

Il faut aussi s'arrimer au modèle sur les acteurs, non?! :)

3a. Merge those kinds of fields?

Je pense qu'il faudrait avoir une idée de tous les champs qui mènent à des E55_Type. Je ne crois pas qu'on doivent les fusionner afin de ne pas créer un modèle P2 > E55. À quelles questions voulons-nous être capables de répondre avec le modèle?

3b. How to manage the classification hierarchy(ies)?

4a et 4b. What is the distinction between the Appellation and the Classification?

4a et 4b. Should we manage the generic object name (e.g. chair) has part of the classification?

Nous avons plus de 11 000 noms d'objets dans notre base de données, mais ils ne sont pas rattachés à une structure classificatoire.

La façon dont on enregistre repose surtout sur la fonction première de l'objet. Donc un veston, même s'il devrait être rattaché à la sous-catégorie Info-Muse "Vêtement de dessus", aura en premier la sous-catégorie "Équipement récréoculturel" s'il a été utilisé comme costume de théâtre, par exemple. Ça demeure vrai que c'est un vêtement, mais ce n'est pas ce que nous cherchons à dire.

Comment conjuguer "fonction/usage" avec les structures classificatoires comme Nomenclature ?

Nommer un objet et le "classer" sont deux choses que nous faisons au catalogage et qui servent surtout pour des besoins administratifs. Parfois, si une recherche a été effectuée par une personne sur une donation, on peut modifier le terme donné par cette personne par un autre terme qui facilite la recherche (ex: piédestal vs. chandelier funéraire)

Aussi, un même objet, comme le flambeau olympique des JO de 1976 aura été "classé" différemment selon s'il est acquis dans un musée de société ou dans un musée de beaux-arts.

Nous sommes habitués de débuter une recherche par un "nom d'objet" (chaise) ou une "classification" (mobilier). Personnellement, j'aimerais qu'on arrive à que ce ne soit plus le point d'entrée pour les usagers.

5a. Should it be part of our current work?

Moi j'aimerais, peut-être pas en premier, mais c'est une information qu'on diffuse sur Collections en ligne. Ça permettrait aussi de faire le lien avec les publications associées à ces expositions, de documenter la provenance, de faire des liens entre les institutions, etc.

6a et 6b. Should we document this in other more structured fields?

Qu'est-ce que "l'art" - ou n'importe quelle autre catégorie/domaine ? Je crois qu'il faut remettre en question ces catégorisations. Elle témoignent plus de la façon dont l'objet a été organisé dans une collection à un moment x et moins de son contexte ou sens.

7a. Should we infer the Period using the date and the place? How to handle the different understanding of the period boundaries (e.g. Victorian Era or Meiji era)? 7b. Do we need to document the period's boundaries?

Il y a juste Fuzzy boundaries + Incertitude qui me viennent en tête :|

7c. Distinction between different beginning events (e.g. when the picture was taken, when the picture was developed, etc.)

Je crois que oui, si nous avons l'information. Aide à mieux comprendre le processus de production, les actions posées. Il y a aussi toute la temporalité associée aux techniques de fabrication : le tissu a été découpé, ensuite cousu, etc. Il faut s'assurer de bien comprendre les propriétés CIDOC et de mon souvenir de la Mesure 113, ce n'était pas si simple.

[PDF] Temporal Primitives, an Alternative to Allen Operators | Semantic Scholar

Temporal hints in the cultural heritage discourse: what can an ontology of time as it is worded reveal? - UMR 3495 MAP

7d. How to handle a label like "First quarter of the 18th century"?

Ou "Début du 18e siècle", "Milieu du 19e siècle", "Fin du 20e siècle"

Je crois que dans ces cas, les institutions ont des définitions : par exemple, "milieu" peut s'appliquer à 1925 po-1975 an (ce qui évite d'écrire 1er quart, 2e quart et 3e quart | ou 1re moitié et 3e quart). Chez nous, par contre, ce n'est pas tranché au couteau.

8a. How to manage "in the style of"?

Et les "d'après", "attribué à" et "?" (incertitude)

Aussi :

Il y a encore une question de temporalité

9.ajout :

Comment gérer les " [...] " et Rio[p]elle. Nous ne pouvons pas toujours retranscrire les signatures (ex : syllabaire inuktituk, alphabet cyrillique, etc.).

À ce sujet, voici la règle de la SMQ : Pour les zones relatives aux inscriptions, telles que Signature et Transcription / translittération, il arrive parfois que l’information ou une partie de celle-ci soit illisible ou difficilement décodable. Si l’information est complètement illisible, ajoutez à l’endroit correspondant trois points de suspension entre crochets […]. Si une partie de l’information est illisible mais que, par déduction, vous pouvez compléter l’information manquante, inscrivez cette information entre crochets [xyz].

10a. Is there a clear distinction between both?

Non. C'est plutôt une règle de saisie. "Matériaux" = 3d ; "Médium" et "Support" = 2D Il demeure que c'est un matériau qui est appliqué sur une surface qu'on qualifie de "support"

Par contre, nous avons des matériaux qui n'en sont pas nécessairement, mais qui sont plutôt le résultat d'une transformation. Par exemple : "Étoffe, feutre" (quand on ne peut distinguer s'il s'agit de laine ou de fibre synthétique) ; "Aliment transformé, gâteau" (nous avons un gâteau de mariage dans les collections) ; "Objets trouvés" (pour les items de ce type.)

Il y a aussi les "matériaux" de nature biologique (poil, humain ; patte, canard ; pierre de vessie) ou encore la drogue et les médicaments. Certaines installations d'art contemporain peuvent avoir le nom des éléments comme matériaux (télévision, câbles, imprimante).

Une bonne définition de ce qu'est un matériau me semble nécessaire afin que les institutions puisse faire les distinctions qui s'imposent.

10b. Is there a clear distinction between medium and technique? (e.g. Albumen Process)

Pour nous, généralement, oui.

Quelques exemples :

10c. Is it possible to distinguish them as they are often recorded in the same field with "Techniques"?

Pour nous, c'est dans des champs séparés. Par contre, on peut avoir médium : Peinture, huile ; Support : Toile et la technique sera simplement "Peint" et non "Peinture à l'huile" Ne pas oublier qu'il y a beaucoup de passés composés dans les techniques. Avant tout, de quoi parle-t-on quand on parle de technique? C'est possiblement mieux détaillé pour les beaux-arts, mais pas nécessairement pour les collections ethno-historiques.

10d. How to link the materials and the techniques?

Ce lien n'est pas mis en évidence dans notre base de données. Revoir le document de travail que Frédéric Bricaud avait commencé concernant l'aquatinte.

Christian-McCord commented 2 years ago

Tous les champs listés ici font partie de nos fiches Object. Étant donné mon manque de familiarité avec CIDOC et la modélisation, je ne peux commenter leur conceptualisation. Voici alors un bref aperçu de comment le Musée McCord les utilise pour le catalogage de ses collections ainsi que pour ses besoins de gestion. La fiche Object est le pivot central d’un système de relations avec d’autres modules et sous-modules : Constituents (Actors), Exhibitions, Medias, Accessioning, Conservation, etc., pour nommer les plus utilisés.

2. Collection a. Collection title Tous les enregistrements d’objets précisent le nom officiel de la collection à laquelle ils appartiennent (champ Department). La collection est identifiée au niveau de la subdivision qui est en soi une classification. Ex: Material Culture: -Material Culture - Commercial and Industrial Objects -Material Culture - Domestic Objects -Material Culture - Military Objects and Weapons -Material Culture - Objects of Science and Technology -Material Culture - Recreational Objects -Material Culture - Religious Objects b. Collection Description Les collections ne sont pas décrites dans la base de données autrement que par leur nom. Elles correspondent aux objets répartis sous la responsabilité de divers conservateurs en fonction de leur domaine d’expertise. À des fins de diffusion, les conservateurs complètent, corrigent et bonifient les fiches Object. Le champ Curator Approved est alors coché, le nom du conservateur est précisé dans le champ Curator et la date de révision dans le champ Curator Revision Date. c. Collections in Collections Des collections au sein des Departments sont identifiées avec des Actors (acquisition-related constituents, role : Collection). On retrouve de grandes collections institutionnelles nommées en fonction de leur provenance (McGill, McCord, Stewart, Lake St. Louis Historical Society et Musée de la Mode), et occasionnellement des collections privées. Cette dernière information peut également être précisée dans le champ Credit Line (ex : Samuel Bronfman Collection for Seagram - La Collection Samuel Bronfman pour Seagram) d. Beginning and the end of the collection (could be computable) Certaines de nos collections sont mortes et ne seront plus augmentées, mais ce n’est pas précisé dans la bd.

Questions: a. What could be part of a collection? Tous les objets physiques et virtuels officiellement acquis qui ont reçu un numéro d’inventaire (NAC). b. Should we focus only on physical things or intangible heritage are also considered as collection items? Nous avons récemment commencé à développer une collection de photographies numériques que nous décrivons de la même manière que les photographies analogiques à l’exception des champs liés à la matérialité des images. Les témoignages ou récits propres au patrimoine immatériel, que ce soit sous formes de transcriptions ou de médias audios et vidéos, ne sont pas actuellement versés dans la bd. c. Should we document the related actors like a donor, a creator, etc.? Absolument. Ce sont des informations cruciales pour une bonne compréhension des objets. Pour le donateur, il faut utiliser le champ Credit Line qui fait état de la façon qu’il souhaite que son don soit crédité.

3. Object Classification/Type

Questions: a. Merge those kinds of fields? Non, je vois mal comment cela peut être fait avec nos données. La classification est une catégorie d’ordre générique, le type (je comprends ici le champ Object Type) est une information spécifique assez pointue qui ne concerne que l’objet auquel il est associé. b. How to manage the classification hierarchy(ies)? Nous référons au site Nomenclature.

4. Object Appellation a. Title name: s’il est ici question du « titre » de l’objet, c’est le champ que nous utilisons pour décrire les œuvres iconographiques (titre et sujets), les archives et les objets de culture matérielle dans le cas de sujets représentés (dans l’esprit du Thésaurus Garnier). b. Object name: est établi en général selon le site Nomenclature. Dans bien des cas, une complémentarité existe avec l’Object Type (ex : Robe de soirée : OB : Dress, OT : Evening). c. Generic object name : c’est pour nous l’Object Name. D’autres appellations sont utilisées : Alternate Object Name, Complete Object Name et Specific Object Name. Nous ajoutons également les appellations diffusées dans notre base de données en ligne. Il s’agit, dans la plupart des cas, de concaténations OB/OT, dans les deux langues. d. Different titles depending on the context (exhibition, language, etc.) : nous utilisons plus de 40 types de titres selon la nature et les besoins des collections cataloguées (ex : artist’s title, caption title, constructed title, inscribed title, published title, etc.) Certains servent à identifier les sujets représentés à l’aide de mots-clés. D’autres relèvent des RDDA (pour les archives).

Questions: a. What is the distinction between the Appellation and the Classification? L’appellation est l’identification de l’objet, la classification précise sa fonction b. Should we manage the generic object name (e.g. chair) has part of the classification? Quelque part le nom de l’objet est une extension de sa classification. Mais les diverses classifications et sous-classifications associées peuvent compliquer les relations entre les deux champs. Ainsi, “Personal Objects-Clothing-Outerwear”, qui est une concaténation d’une classification et de deux sous-classifications, renvoi dans notre bd à près de 150 noms d’objets distincts. Les mêmes noms d’objet peuvent être utilisés dans d’autres classifications et collections avec un sens différent.

5. Exhibition

Questions: a. Should it be part of our current work? Nous utilisons le champ Exhibition history pour nommer les expositions internes et externes (avec les dates et les lieux de présentation) qui ont inclus l’objet. Une relation avec la fiche Exhibition permet d’accéder à toute une gamme d’autres informations concernant la production de nos propres expositions.

6. Subject a. Depicted subject(s) b. Categorization aspect (e.g. Art)

Questions: a. Should we document this in other more structured fields? Concernant le champ Subject (sujets représentés), nous utilisons un système de descripteurs ou mots clés de différents niveaux, allant du plus générique au plus spécifique. La structure de ce système permettait des macro recherches à une époque où celle-ci s’effectuait par chaînes de caractères. Cette structure n’est plus nécessaire avec les logiciels qui utilisent des opérateurs logiques. Nous la maintenons surtout pour s’assurer que nous avons couvert tous les aspects possibles de la description du sujet. Le champ Subject cohabite avec les divers champs Titles où on retrouve souvent les mêmes termes (noms, lieux, etc.). Ce champ est peu utilisé pour les collections d’objets 3D et de vêtements.

7. Date and Period

Questions: a. Should we infer the Period using the date and the place? How to handle the different understanding of the period boundaries (e.g. Victorian Era or Meiji era)?
Le champ Period est essentiellement utilisé pour identifier un siècle ou une partie d’un siècle (quart, tiers, moitié). Pour la collection archéologique, on y retrouve également des noms de civilisations historiques (ex: Paleo-Eskimo, Thule, etc.). b. Do we need to document the period's boundaries? Risqué, les limites temporelles peuvent varier d’une institution à l’autre… c. Distinction between different beginning events (e.g. when the picture was taken, when the picture was developed, etc.) Pour les œuvres iconographiques, la date historique du sujet (évènement représenté) est précisée dans le titre; la date de production de l’objet se retrouve dans le champ Date. d. How to handle a label like "First quarter of the 18th century"? Dans ce cas précis, les champs Begin Date et End Date sont 1800 et 1825. Le champ Date est 1800-1825 et la formulation entière est affichée dans le champ Period.

8. Related Actors a. Specific roles in different contexts (donors, creators, artists, etc.) Nous utilisons 80 types de rôles pour les créateurs des objets (Object-related constituents). Les termes sont validés avec la hiérarchie « Makers » du AAT.

Questions: a. How to manage "in the style of"? Cette information est spécifiée dans le champ Prefix du Display Name de l’Actor artiste auquel elle réfère: « in the style of Cornelius Krieghoff ». Le champ Suffix permet d’ajouter un « ? » ou autre s’il y a lieu. Le Display Name de la fiche Object permet aussi d’afficher une appellation alternative de l’Actor, et ses dates biographiques. Les Alternate Names de l’Actor sont gérées dans la fiche Constituent.

9. Signature a. Physical marks description (color, etc.) Ex : Handwritten in pencil b. Localisation Ex : abbréviations (l.l. within image, u.r corner, l.r. below image, etc.) c. Signature text a., b. et c. sont précisés dans le même texte, et, dans la plupart des cas, confirment les noms des Actors liés. Champ Signed pour les œuvres 2D, champ Mark(s) or Label(s) pour les objets 3D et vêtements.

10. Material, Medium and Support Ces 3 types d’informations cohabitent dans le champ Medium / Technique. Le support est également décrit dans le champ Paper/Support. On les retrouve aussi sous la forme d’attributs gérés dans un thésaurus d’autorités développé localement. a. Information (artwork, etc.) material b. Support material (original and added by the curators) Les supports ajoutés par les restaurateurs sont précisés dans les fiches du sous-module Conservation (rapports de traitement).

Questions: a. Is there a clear distinction between both? Avec les attributs (autorités appartenant à différentes catégories) = oui. Avec le champ Medium / Technique (énumération des mediums et supports dans un texte) = non. b. Is there a clear distinction between medium and technique? (e.g. Albumen Process) Dans le champ Medium / Technique la technique est isolée sur une ligne à part avec un retour de chariot. Dans le cas d’un papier albuminé, les données se déclinent ainsi : Object Name = Photograph Specific Object Name= Albumen print Medium / Technique = Silver salts Albumen process Paper/Support = Paper mounted on paper Attribute = Albumen Process (type: Technique) c. Is it possible to distinguish them as they are often recorded in the same field with "Techniques"? Selon le type d’attribut. d. How to link the materials and the techniques? Ce sont deux informations distinctes que nous énumérons dans le champ Medium / Technique dans un format qui se rapprochent de celui des cartels d’exposition.

11. Techniques a. used in the production to achieve the final product Notre thésaurus comprend près de 1200 termes pour le type Technique. Plusieurs attributs Technique peuvent être reliés à une même fiche Object

12. Place of production a. could also be place of origin of object En effet. Dans le sous-module Geography, nous localisons les objets avec 4 types de descripteurs géographiques: Borden Code, Geocultural, Origin, Used

Questions: a. Would it be enough to reused only a place URI or sometimes it is useful to document the "place within a place" hierarchy (eg. Montreal, Quebec, Canada, North America) Selon la nature de la collection, des hiérarchies de localisations sont précisées. Elles sont en général d’ordre politique, mais physique pour les collections autochtones et archéologiques:

image

b. How to handle historical places? Nous utilisons les toponymes historiques et précisons leurs équivalences contemporaines entre parenthèses.

13. Description of object (physical) a. dimensions: width, height, depth, weight, size (when open/close) = Display Dimensions: système métrique, = Dimensions Assistant: systèmes métrique et Impérial b. shape, color, scent, texture, etc. = Attributs c. distinguish values from unit of measurement, e.g. 2 (value) metre (unit) Dans notre système, “metre” est un Dimension Type, Primary Unit correspond au système métrique, Secondary Unit, correspond au système impérial d. condition: physical description or its own pattern? La condition de l’objet est décrite dans le sous-module Conservation dans une fiche en soi (examinateur, date d’examen, type d’examen, remarques, médias, etc…)

14. Description of object (textual) Champ Description. La description tend à être plus détaillée en l’absence d’une image.

15. Supporting texts a. Notes/remarks/research/bibliography Nous avons les champs textes: Notes / Label Text (cartels d’exposition) / Curatorial Remarks / Historical Attributions / Research References / Related Works

16. Partitioning of object a. an object can be composed from multiple components, e.g. a tea set includes a tea pot, cups and saucers b. each part itself can be described as a standalone object c. linking parts to whole and vice versa d. Number of components Les composantes sont gérées dans le sous-module Components. Plusieurs champs sont disponibles. Nous utilisons essentiellement les champs suivants : _Component Number, Component Name, Count et Component Type_

Questions: a. Do we need to make a difference between a piece that you can remove from a part that you cannot? Selon notre procédurier, la pièce qui peut être retirée d’un objet est une composante.

17. Location a. current display location and/or storage b. physical and/or digital (including object's URL?) Pour des raisons de sécurité, les localisations ne devraient pas faire partie du modèle.

18. Provenance a. type of transaction, date, place, parties involved (doner, buyer, seller, auction house, broker, etc.), monetary value (if applicable), identifier of transaction. Ces informations, confidentielles, se retrouvent en partie avec les Acquisition Related Constituents et dans le sous-module Accessioning. Elles ne devraient pas faire partie du modèle.

Questions: a. only its provenance prior to being acquired by data provider, e.g. donation by..., purchase from...; or an object's entire provenance if documented and available? Ces informations sont versées dans le champ Provenance.

19. Image a. Image of object b. URL

20. Rights a. of images b. of object itself La gestion des droits sur les images est effectuée dans le sous-module Rights and Reproduction, qui, à l’heure actuelle, est à jour seulement pour les images disponibles sur notre base de données en ligne.

21. Style/Culture Nous n’utilisons pas le champ Style. Le champ Culture est utilisé seulement pour la collection autochtone.

22. Publications a. Article where the item has been mentioned Champ Published Object References

Habennin commented 2 years ago
  1. Identifier a. Manage identifiers from different provenances

reference to an E33 as source or an E13 as attribute assignment, make it a basic pattern.

b. Shared identifiers for different components

give the identifier a URI and use the same URI and class for both, probably have an E13.

  1. Collection

E78

Even when one is handling intangible heritage, typically one is mangaging a real item which forms the basis for the building of knowledge around an intangible heritage practice. One can collect things, one can't collect the intangible (in the sense of putting it in a box, on a shelf and looking after it... one hopes). So this doesn't mean exclude or do not value intangible heritage, it means it is not a physical object and so should be treated under another facet!

a. Collection title

E41

b. Collection Description

E33

c. Collections in Collections

P46

d. Beginning and the end of the collection (could be computable)

E87 ->p4 ->E52 ...

Questions: a. What could be part of a collection?

physical things (that might reference or be the historical connection to important intangible heritage)

b. Should we focus only on physical things or intangible heritage are also considered as collection items?

Could you consider a song or a dance part of the collection? What would that mean? Does the museum own it?

c. Should we document the related actors like a donor, a creator, etc.?

Seems like it would enrich the knowledge around the object.

  1. Object Classification/Type

Questions: a. Merge those kinds of fields? b. How to manage the classification hierarchy(ies)?

Object classification and type are often very different. Usually 'type' in the strict sense is a form breakdown of the object. Classification is often a functional breakdown. Best to keep separate and use metatypes. (even suggest full types to CRM SIG?)

  1. Object Appellation a. Title name b. Object name c. Generic object name

All E41_E33 with types!

d. Different titles depending on the context (exhibition, language, etc.)

E13 ->p20 -. E7 or the long awaited CRM Soc classes :/

Questions: a. What is the distinction between the Appellation and the Classification? b. Should we manage the generic object name (e.g. chair) has part of the classification?

Often times the term 'name' is used especially in emu systems to point to type. It should be mapped as a type imho.

  1. Exhibition

Questions: a. Should it be part of our current work?

It's up to you, but it's very rich as a contextual event. Interesting to see the object's contact with people and places over time? Although that would suggest an event/exhibition facet perhaps.

  1. Subject a. Depicted subject(s)

big CRM question here is if you want to make a visual item for every physical item or just use the shortcut

b. Categorization aspect (e.g. Art)

?

Questions: a. Should we document this in other more structured fields?

  1. Date and Period

Questions: a. Should we infer the Period using the date and the place? How to handle the different understanding of the period boundaries (e.g. Victorian Era or Meiji era)?

If you wanted to do it, you could always add some provenance nodes in order to indicate that this was an algorithm opinion and not a human based one.

b. Do we need to document the period's boundaries?

It puts a lot of curatorial burden. Where would you draw the lines of what you wanted to cover and not (geogrphically and temporally) You have seen chronontology? https://chronontology.dainst.org/

c. Distinction between different beginning events (e.g. when the picture was taken, when the picture was developed, etc.) d. How to handle a label like "First quarter of the 18th century"?

You can just create a label for the timespan and leave it at that. You could provide instructions to provider institutions on potential useful translations (e.g.: for this kind of data make a start and end that begins with the 00 and goes to 25)

  1. Related Actors a. Specific roles in different contexts (donors, creators, artists, etc.)

PC14!

Questions: a. How to manage "in the style of"?

does it point to a person... you can point it to that person from the E12 node using influenced by P15 OR if it points to a type then you could use P32 used general technique

  1. Signature a. Physical marks description (color, etc.) b. Localisation c. Signature text

If you really wanted to make this traceable then make an instance of E25 HMPP and then you can have marks description in an E33 that references it (P67i), the signature can carry p128 the text (E33) or the signature (p2) E55, for localisation there is often just more free form text. There are some terms in AAT for upper right, lower corner and such as well in case you wanted to chase standarization.

  1. Material, Medium and Support a. Information (artwork, etc.) material

Often times this is mixed. The linked.art solution of just having a 'material and medium' description using E33 seems adequate. Just in case the museum has more analytic data, you can use it.

b. Support material (original and added by the curators)

if you had rich information about this then you could make another instance of E22 HMO and then use standard patterns.

Questions: a. Is there a clear distinction between both? b. Is there a clear distinction between medium and technique? (e.g. Albumen Process) c. Is it possible to distinguish them as they are often recorded in the same field with "Techniques"? d. How to link the materials and the techniques?

It depends on the institution. Usually not. You can use the mixed descriptions to isolate terms and then split them. One links the material and technique often via the E12 production node. This production used this material and this technique to make this object.

  1. Techniques a. used in the production to achieve the final product

P126 employed

  1. Place of production a. could also be place of origin of object

Then it isn't the place of production! Should create a new event node cluster and use the encounter event class from CRMSci.

Questions: a. Would it be enough to reused only a place URI or sometimes it is useful to document the "place within a place" hierarchy (eg. Montreal, Quebec, Canada, North America)

If it uses a well known gazeteer then you get this by inference?

b. How to handle historical places?

Gazeteer?

  1. Description of object (physical) a. dimensions: width, height, depth, weight, size (when open/close)

just E54, pretty well handled by CRM. Open and close can be handled by indicating the method of measuring on the measuring event.

b. shape, color, scent, texture, etc.

type? or description?

c. distinguish values from unit of measurement, e.g. 2 (value) metre (unit) d. condition: physical description or its own pattern?

Might want to keep the E3 pattern in order to have maneuverability in the future. Condition changes after all.

  1. Description of object (textual)

E33

  1. Supporting texts a. Notes/remarks/research/bibliography

E33 and type... but you may want a textual work facet, since this will not be unique to the individual object, likely.

  1. Partitioning of object a. an object can be composed from multiple components, e.g. a tea set includes a tea pot, cups and saucers

P46

b. each part itself can be described as a standalone object

question of granularity of documentation

c. linking parts to whole and vice versa

if you are thinking based RDF XML then this is a non problem. You get bidirectionality out of the box

d. Number of components

Maybe better use dimension here than the really old property.

Questions: a. Do we need to make a difference between a piece that you can remove from a part that you cannot?

  1. Location

use a move pattern e8

a. current display location and/or storage b. physical and/or digital (including object's URL?)

  1. Provenance a. type of transaction, date, place, parties involved (doner, buyer, seller, auction house, broker, etc.), monetary value (if applicable), identifier of transaction.

usually shows up as a text. If so, just make it an E33. If you have actual modules, then you embark on modelling provenance, probably another facet (can reference linked art modelling)

Questions: a. only its provenance prior to being acquired by data provider, e.g. donation by..., purchase from...; or an object's entire provenance if documented and available?

Why would it not be all? That's what one looks for in provenance

  1. Image a. Image of object b. URL
  2. Rights a. of images b. of object itself
  3. Style/Culture
  4. Publications a. Article where the item has been mentioned

1-4 should be something that could be standardized across multiple facets over time. 3 seems like a classification question.

So those are my two cents. GB

marielmat commented 2 years ago

En vue de la 13e rencontre du comité sémantique, voici mes premières réflexions sur les points 12 à 18 résumés par @illip

12. Place of production
a. could also be place of origin of object

Effectivement. Il y a des institutions qui font la différence entre le lieu de production et l'emplacement d'origine. Cette distinction est aujourd'hui dans les règles Info-Muse, mais elle ne l'a pas toujours été. Un objet fabriqué en Chine mais acheté aux États-Unis pourrait avoir comme "Pays d'origine" (le champs Info-Muse original) l'un ou l'autre des deux pays.

De notre côté, quand on consigne une nouvelle institution dans notre base de données, on essaye d'ajouter une adresse. On peut donc soutirer le "Pays du fabricant" de là.

Pour moi, ça serait important de partir de la production et de trouver comment faire valoir que mon objet a été fabriqué en Chine, même si le siège social était ailleurs, puis qu'il a été vendu au magasin x tel endroit, puis acheté par telle personne pour l'offrir à son enfant, aujourd'hui notre donateur, un Noël de son enfance.

12a. Would it be enough to reused only a place URI or sometimes it is useful to document the "place within a place" hierarchy (eg. Montreal, Quebec, Canada, North America)
12b. How to handle historical places?

Lien avec culture et les aires culturelles. Je ne crois pas que seul le URI soit suffisant. Je crois qu'une partie dépend de ce que vous visez comme visualisation. Je vais tenter de trouver des exemples ...

13a. dimensions: width, height, depth, weight, size (when open/close)

D'accord avec le commentaire de @habennin.

Je rajoute certains types : dimensions du cadre, dimensions de l'image, dimensions de la base, poids = lien avec les parties composantes.

Penser aussi que les gens n'utilisent pas uniformément les termes (ex : selon Info-Muse, on parle de "hauteur" pour un pantalon, mais dans le langage courant, plusieurs parlent de "longueur" = on a les 2 dans notre bd.) = La façon dont on consigne les dimensions dépend beaucoup de la façon dont nous percevons l'objet.

13b. shape, color, scent, texture, etc.

D'accord avec @Christian-McCord. Principaux éléments descriptifs sont consignés en libre saisie.

Avons déjà eu des champs précis (ex : motif ou couleur - en Attributs, comme Christian), mais c'était très subjectif (qu'est-ce que "kaki" ?) Je rajouterais : la disposition des éléments les uns par rapport aux autres. J'ai l'impression que certains de ces éléments ont un lien avec la modélisation des images qui sont associées aux objets.

13d. condition: physical description or its own pattern?

Cette donnée n'est généralement pas diffusée.

Çe me ramène aux Techniques de fabrication. "Cousu", n'est-ce pas un "état" dans lequel se retrouve une matière après x intervention? Tout comme les termes qu'on utilise pour décrire les états : "Corrosion" ou "Piqûre", par exemple.

14.Description of object (textual) 15. Supporting texts 15a. Notes/remarks/research/bibliography 22.Publications

Transcriptions et signatures ? + Emplacement ;)

Comment extirper la richesse des informations consignées dans les champs texte ? - souvent intéressante pour remettre l'objet en contexte.

Comme on documente les objets à la pièce, plusieurs ont souvent le même ou du moins une partie du même "commentaire historique".

Pour bibliographie > possibilité d'arrimer ces documents avec FRBRoo (LRMoo?). De notre côté, deux types de références : Publications (dans lesquelles on retrouve nos collections) et Références documentaires (qui ont servi à bonifier la fiche). Je vois peut-être un lien avec Provenance et Expositions.

16a. an object can be composed from multiple components, e.g. a tea set includes a tea pot, cups and saucers 16b. each part itself can be described as a standalone object
16c. linking parts to whole and vice versa
16d. Number of components

Pour moi, le service à thé n'est pas un objet mais un concept : telles et telles choses, qui peuvent être formées d'autres éléments comme un "couvercle", forment un ensemble qu'on nomme "Service à thé".

Besoin de tout concevoir comme des objets? Quand savoir qu'un "composant" entre dans la composition de quelque chose de plus grand (ex: couvercle + récipient = sucrier) ou ne fait que l'accompagner (ex: étui quelconque, socle)

Distinguer les objets identiques ? Les paires?

Besoin de comprendre la façon dont les objets sont numérotés.

17a. Location | current display location and/or storage

storage = généralement confidentiel

intéressant pour la mise en valeur des expositions et des liens entre les institutions.

17b. physical and/or digital (including object's URL?)

Voir 15a. Emplacement des signatures, transcriptions, etc. - lien avec la localisation de ces éléments sur les images

18a. only its provenance prior to being acquired by data provider, e.g. donation by..., purchase from...; or an object's entire provenance if documented and available?

Certaines informations sont confidentielles. Le concept de Provenance est néanmoins très important pour certains musées et mérite d'être diffusé. Ça rajouterait un + de s'y attarder un peu, à mon avis (même si je trouve la proposition de @Habennin plus réaliste pour maintenant)

Un groupe a été formé en 2020 pour proposer à GallerySystems (fournisseur de TMS) des amélioration quant à la gestion de la provenance dans TMS. Par ailleurs, GallerySystems a lancé un sondage cette semaine à ces usagers pour sonder leurs besoins en ce sens.