Structuring and modeling questions - archival boxes and similar packaging

[Arian Rajh]

Dear all, 

So, 2 questions. The first one refers to structuring. The second one is connected, but relates to modelling common information for us. 

I have some considerations on how to express a particular series consisting of 12 boxes of archival materials. The content of these boxes is record sets on file level and records (item level), and thinking about the following modelling solution:

1. Describe the whole series as a top-level record set

2. Describe file and item-level record sets and records

3. Describe their instantiations (1 instantiation per document-item, and 1 instantiation per file-record set, as both RiC-O and RIC-AG allow record sets to have instantiations too)

4. Add information about the box to each instantiation. I am considering two options here: the first is that boxes are places of storage, so I could model them as cases according to the terminology in the software I'm using. The second option is to model archival boxes as instances of a class, equip them with identifiers, and link them to instances of records and record sets. I initially considered the CarrerType class, but based on its description in RiC-O and RiC-A05 in CM (from which it derived), this doesn't seem like the correct solution.

Second question: How do you express information about the archival box in which some material is stored? What would be the best way? I have searched our group for topics on this matter, but have had no luck, so I would appreciate the community's advice.

Kindly,

Arian

[Florence Clavaud]

Dear Arian,

Just my two first cents about yoour questions:

- yes a series,  files and items can definitely be considered instances of rico:RecordSet (as concerns series and files as you describe them, i.e. as ISAD(G) defines them, or as defined there by RiC-O, using SKOS concepts, instances of rico:RecordSetType:  https://www.ica.org/standards/RiC/RiC-O_1-1.html#Series; and https://www.ica.org/standards/RiC/RiC-O_1-1.html#File) and of rico:Record

- yes you can describe their instantiations, which would for example have an extent, identifiers (like reference codes, shelfmarks, etc.), a carrier type, the institution which holds them, etc. 

- about boxes: I would tend to consider them as instances of a class, which may help you assign them identifiers (barcodes for example), a location if  needed, etc. 
RiC-O does not have yet any specific class for this yet, but you could extend it and create a subclass of rico:Thing.
As concerns how to extend RiC, in case you would like to know how you can do this, see this FAQ in RiC-AG, which includes a small example about extending RiC-O: https://ica-egad.github.io/RiC-AG/faq--how_to_extend_ric.html. 

About this topic, see also this thread of the RiC users list, starting from a question asked by Aaron Hope in 2024: https://groups.google.com/g/records_in_contexts_users/c/eB2zWCeS55o/m/ljh_4l20DQAJ

Hope this helps,

Florence Clavaud
Chair of EGAD, lead of RiC-O development team

[Arian Rajh]

Dear Florence,

Thank you very much for your advice! If I may add, I think it would be beneficial to clarify whether to address archival box and similar emballage through a new or some expanded class (options A and B in RiC-O), or instead establish a convention for handling boxes and emballage in RiC-AG (option C such as modelling them as storage place elements). This information about archival boxes comes up often, so this would help unify our practice. 

Kindly,

Arian

[Florence Clavaud]

Dear Arian,

We may consider adding a new rico:Container (or something of the kind) class in RiC-O. Meanwhile, you still can extend the ontology as I suggested.

Best regards,

Florence Clavaud

[Johan Pieterse]

Hi Adrian
Your modelling approach is valid. I would treat the archival box as a distinct container entity rather than only as a storage place, because it can have its own identifier, barcode, and changing location over time. In practice, I would model the series/files/items as Record Sets and Records, model their instantiations as needed, and then link those instantiations to a locally defined Box class, perhaps as a subclass of rico:Thing, while linking the box separately to its storage location. That seems cleaner than forcing boxes into an existing RiC class that does not fully fit.  

[Arian Rajh]

Sounds great, thank you,

Arian

Dr. sc. Arian Rajh, izv. prof. 

Poslano s mojeg iPhonea

31.03.2026., u 17:42, korisnik Florence Clavaud <florence...@culture.gouv.fr> je napisao:



--
You received this message because you are subscribed to the Google Groups "Records_in_Contexts_users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to Records_in_Context...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/Records_in_Contexts_users/d01f8928-5e0b-4ec0-955d-5ddb1f858f0cn%40googlegroups.com.

[Arian Rajh]

Dear Johan, thank you for the advice - yes, I agree as it makes more sense.

Kindly,

Arian

--
You received this message because you are subscribed to the Google Groups "Records_in_Contexts_users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to Records_in_Context...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/Records_in_Contexts_users/06502465-a011-4dd3-b519-20d16b39384dn%40googlegroups.com.