Release of RiC Application Guidelines (RiC-AG) v0.1

[CLAVAUD Florence]

Dear RiC users,

After months of intensive work, the EGAD group is delighted to announce the publication of version 0.1 of RiC Application Guidelines (RiC-AG).

RiC-AG is currently available online as a website, accessible at https://ica-egad.github.io/RiC-AG/. We will also be publishing a PDF version soon.

If you wish to browse the website locally on your computer, you can download the release at : https://github.com/ICA-EGAD/RiC-AG/releases/tag/v0.1.0. Once download and unzipped, you will find the website in the 'docs' subfolder (the home page is the index.html page). 

RiC-AG is the last of the four complementary documents that constitute the Records in Contexts standard and address the activity of describing records. It is intended to provide archival practitioners, developers, and managers guidance in understanding and implementing archival description systems based on RiC, particularly focusing on the RiC Conceptual Model (RiC-CM).

When designing the AG, EGAD took into account the comments received on RiC-CM 2.0, the user stories gathered in 2024, feedback from users who already have implemented RiC, and many discussions held with members of the concerned communities. Nevertheless, while RiC-AG aims to provide broad guidance that will be helpful across a range of scenarios, it does not offer specific guidance that will address every possible user context.

RiC-AG is a draft document. With the initial release, the EGAD invites the community to provide feedback.

RiC-AG is dynamically generated from sources managed on GitHub. The EGAD group will therefore be able to develop it step by step, whether it involves modifying details, adding new sections or FAQs.

Best regards,

Florence Clavaud,
Head of the Lab, Archives nationales de France
Chair of EGAD

---

Merci de nous aider à préserver l'environnement en n'imprimant ce courriel et les documents joints que si nécessaire.

[Jochen Deprez]

Hello,

I just did a full read through, very well-made and clear guidelines to get started! I really like the example given in the "Getting started" chapter, especially the link with RiC-O. The "form" approach and then the final graph view is a good way to represent the data entered and the final result. 

What i feel is still missing though, are the following:

1. The example represented in Excel, since this is a tool many archivists use in order to store and prepare data.
2. A full RiC-O example (so the same thing, but in RiC-O).
3. A full example in turtle-triples and/or XML (so the same thing again, but now in turtle or XML, so we can import this in software).
4. Guidelines on how to visualise the data using software like GraphDB, Jena, ...
5. Guidelines on how to set up a SPARQL search engine.
6. Guidelines on how to translate this to a user-friendly web application front end so the public can research the data.

This way, we can full grasp the concept and the possible approaches into implementing it, step-by-step, both full-scale with all bells and whistles, but also more ad-hoc and "quick and dirty". I once saw a demonstration of Ivo Zandhuis (City Archives of Amsterdam) about implementing RiC-O with Excel, i'd like to see this in a formal guideline if possible.

Another question: The note-taking app Obsidian makes it possible to "link" notes and generate a graph view. Would this be usable with RiC, as a way to get a feel for it? 

Greetings,

Jochen Deprez

Archivist and information manager DVV Midwest, Belgium

Op vr 31 okt 2025 om 12:06 schreef CLAVAUD Florence <florence...@culture.gouv.fr>:

--
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/PR0P264MB3289D992BBC7389D7CB106F0D5F8A%40PR0P264MB3289.FRAP264.PROD.OUTLOOK.COM.

[Hope, Aaron (MPBSDP)]

Dear Florence and EGAD,

Thank you for sharing the much anticipated RiC-AG. I have been reviewing it over the last couple weeks and am impressed at the work that has gone into this resource. I found it helpful in confirming some of our modelling decisions and it also offers a lot of very interesting possibilities for future development that I will ponder over the long-term.  A few quick comments and observations:

1) The Record described in section 6c.3 (“When does a derived Instantiation represent a new Record?”) which is re-used as a researcher’s source material and/or evidence in a criminal investigation and thereby becomes a new Record: if the informational content of the original Record hasn’t changed, couldn’t one simply position the same Record in the new context(s)? In other words, can’t a single Record be modelled in multiple Record Sets?

2) The possibilities for modelling intricate details of activities, records, instantiations, agents etc. through time are fascinating but also probably beyond the scope of what an archivist can typically do manually. Therefore, I wonder if some future revision of RiC-AG could include a section on automation of descriptive efforts?

3) Another possible area for guidance in future might be considerations when designing a structure for URIs/IRIs: whether to embed human-readable meaning or not, how to generate unique values, when to use existing external URIs vs internal ones etc. etc.

Anyway, my thanks again to Florence, Richard and the whole EGAD team.

Regards,
Aaron Hope

 

From: records_in_c...@googlegroups.com <records_in_c...@googlegroups.com> On Behalf Of CLAVAUD Florence
Sent: October 31, 2025 7:07 AM
To: 'Google Groups' via Records_in_Contexts_users <records_in_c...@googlegroups.com>
Subject: [Records in Contexts users] Release of RiC Application Guidelines (RiC-AG) v0.1

 

CAUTION -- EXTERNAL E-MAIL - Do not click links or open attachments unless you recognize the sender.

--

[Florence Clavaud]

Dear Aaron,

Sorry for reacting so late, and thank you for this positive feedback!

As concerns 1): yes of course, a single Record can be described, using RiC, as being included in several Record Sets. This can be done easily with RiC-CM, using RiC-R024i 'is or was included in' relation (whioch can connect a Record or Record Set, to a Record Set). RiC-CM makes it really possible and easy to express this fact, which is not the case with ISAD(G) or EAD (with which it is at least difficult, if not impossible, which may often create redundancies), whereas in reality many archival resources can be included in several record sets simultaneously or through time. Same for representing sequences: the same record resource may be involved in different sequences of record resources. See on this topic the FAQ on sequences in RiC-AG (https://ica-egad.github.io/RiC-AG/faq--sequences.html). You can also, of course, do the same in a RiC-O dataset, which may involve more properties. 

As concerns 2): I am not sure I understand exactly what you mean by 'automation of descriptive efforts'. 
If you mean automatically generating a RiC-CM or RiC-O very precise and detailed graph of entities from existing metadata, this would depend on the existing metadata and its structure first, so it is quite difficult to provide generic and interesting advice in the Guidelines about this - except what we have already written IMHO: analyze the data, assess its quality and improve it if necessary, define your needs and expectations, then your priorities, prepare ad hoc mappings and transformation rules, etc.

Ifyou mean generating detailed metadata using AI, from digital or digitized documents, or generating RiC-based metadata from existing metadata using AI, yes we can think of a FAQ, but it would remain quite general also. We could also organize some events about this topic. Anyway having a consistent OWL DL ontology may help a lot, in addition to have vocabularies and authority data.

As concerns 3): about guidance on URIS/IRIS: yes we could try and add some FAQ about this. Like for many other topics, there are various methods, not a single solution; also, some aspects of the topic are not specific to the world of records or archival repositories. However of course it is an important topic. We could start from some existing examples.

Best regards,

Florence Clavaud
chair of ICA/EGAD

[Florence Clavaud]

Dear Jochen

Thank you for your feedback! Just a few initial answers about your comments and suggestions. 

First of all, as we said and wrote, this 0.1 version of RiC-AG focuses mainly on RiC-CM. So though you will find in the guidelines some sections and examples which are about using RiC-O, those remain relatively uncommon.

Of course, we plan to add more about using RiC-O. 

As concern Excel, yes we could provide an Excel version of the Getting Started example. However, everybody should remember that it can become quite complicated to use only Excel to manually prepare metadata in order to generate RiC-O data from this Excel metadata, especially as the amount of data significantly increases, if many different categories of entities are described, and if multiple relationships exist between these entities. In such cases, it becomes necessary to assign an identifier to each entity, and to use the right identifier to point to the description of the target entity in an Excel sheet, from a different Excel sheet that describes the source entity of the relationship (or to rely on the names of the entities, which is quite risky). In case you manually enter all this data, you could make input errors, which could compromise the generation and quality of the output RDF graph. As soon as you have a significant number of entities and relationships to describe, it is more efficient to rely at least on a database.

If you need full RiC-O examples, you already have a lot in the RiC-O repository on GitHub (https://github.com/ICA-EGAD/RiC-O/tree/master/examples). They are also available in the release (https://github.com/ICA-EGAD/RiC-O/releases/tag/v1.1). You can also find links to some datasets in the List of resources on RiC (https://ica-egad.github.io/RiC-ResourceList/). We also invite all those who have produced RiC datasets to reference them using this list.

We plan to publish more examples (and if we can, to make them available through a UI and SPARQL end point). 

Some of your suggestions concern semantic web technologies or tools by themselves, and are not specific to RiC-O datasets. They are addressed elsewhere in multiple training courses, tutorials, software documentation, or specifications. And we need to focus on our main topic, which is RiC, and which already requires a very significant amount of work time. So we may not provide  guidance on these topics in RiC-AG (for example, on how you set up a SPARQL search engine). Rather, we can organize in person training courses where we use open source triplestores like GraphDB, or events where users present on their implementations.

We also may publish a set of generic documented SPARQL queries that anybody could use and adapt, either to query a RiC-O graph or to infer new facts from an existing RiC-O graph; plus a set of generic SHACL shapes which any body could reuse and adapt, in order to check the quality of a RiC-O graph (e.g. any Record Resource should have a rico:title).

best regards,

Florence Clavaud
Chair of ICA/EGAD