RiC translations
Arian Rajh
Florence Clavaud
Dear Florence, Richard, and colleagues from EGAD,We finally got a project (actually an activity in a new project related to disruptive technologies in a cultural heritage environment) as a framework for translating RiC into Croatian. I will work on this with two of my colleagues from the Croatian National Archives (CNA), so two institutions will be involved: the Faculty of Humanities and Social Sciences, Zagreb, and CNA. This is great because it should ensure the translation's broad acceptance among Croatian archivists. My colleagues from CNA (and their co-workers) will translate RiC CM and add a national annex covering some nationally specific situations with additional metadata granularity (E.g., the name attribute A28 from CM is divided into type, value, name part, name part type, date of name, and name status - to illustrate this granularity. The national metadata specification was done when RiC was in version 0.2). Anyway, my colleagues from CNA aim to familiarize the (professional) public with the RiC standard so that they will focus on RiC CM. My task will be to translate RiC-O since my long-term goal is to enhance the utilization of Semantic Web concepts for solutions in my country. My part of the work should be finished by the end of this year.
1. Before starting the work, I would like to know whether I should translate 3 CSV lists of components (classes, data properties, object properties). After translating, will EGAD be able to use these results to update the RiC-O RDF/XML file (by using some scripts)? Or, what would you suggest as an approach to translating RiC?
Yes, as already said, you can use the CSV files released with RiC-O 1.1. Just download them. You can find them either in the release (https://github.com/ICA-EGAD/RiC-O/releases/tag/v1.1), or directly there: https://github.com/ICA-EGAD/RiC-O/tree/master/ontology/current-version/CSV_lists_of_components.
Once you get them, simply open them using a tool like MS Excel or OpenOffice Calc. They files are encoded in UTF-8 and have comma as the value separator, and the column values are enclosed in quotation marks ('"'). Please make sure to keep this encoding format and structure.
Then, do not remove any of the columns in English, but add any column you need for translating RiC-O to Croatian (e.g. a column for the labels in Croatian). Of course, we would recommend you start by translating the English labels of the components. Remember that if you translate everything (or something) to Croatian, you will have to update the translation as RiC-O evolves.
I would also like to draw your attention to another important point: the textual definition (rdfs:comment), the scope note (skos:scopeNote) and, most often, the examples when there are any (skos:example) have the same content in RiC-O 1.1 and in RiC-CM 1.0, when the RiC-O component corresponds exactly to a RiC-CM component (so in particular for all RiC-CM entities, which become classes in RiC-O; for all RiC-CM relations, which become object properties in RiC-O; and for some RiC-CM attributes, in particular those with 'free text' as value type, which become datatype properties in RiC-O). So, your translations of these definitions and scope notes, if you decide to produce them, should be the same for RiC-CM 1.0 and for RiC-O 1.1. Don't do them twice!
Once this is done, please check the translation very carefully (ideally by asking a group of Croatian professionals or some national Croatian institution, like the Croatian State Archives, to validate it), and just transfer the resulting files to us.
We will then be able to easily integrate the CSV formatted Croatian translations into the OWL files, by using a script. I am saying ‘files’, because the modularized version of RiC-O (the version splitted into several files) will most probably become the reference one when we release the next version of RiC-O. So we will probably make the Croatian translation one of the files of this modularized version. We cannot keep a single file growing indefinitely, and managing a collection of files may become much easier. In this future modularized version, some translation file of the English version, in any language (except perhaps for French and Spanish, which are the two other official languages of the ICA), can thus very easily be either used or ignored by RiC-O users, depending on their working language(s).2. Regarding the sustainability of this project's activity and our national particularities, I'd like to know if it is possible to add additional granularity/metadata as a modularized file (this is not part of this project but a potential activity for its sequel). If this is possible, then modularized files will be possible to use as a sort of metadata backbone of our IT archival system someday (I'm just thinking aloud).
This question (or related questions) is often asked and discussed at the moment (see for example this email: https://groups.google.com/g/Records_in_Contexts_users/c/sfb6uAjV8d4/m/UECMwvLjAQAJ) . I will limit myself here to giving an initial general answer, which we will complete later.
You can extend RiC-O as you like, adding subclasses, or sub-properties that would suit your needs (again, more on this soon). However, as these additions would not be as such a part of RiC-O, they should be defined in a distinct ontology that should have its specific URI (e.g. http://www.myInstitution.org/dataModel/ontology). The components added should have their own namespace URI (e.g. http://www.myInstitution.org/dataModel/ontology# ) and preferred namespace prefix (e.g. ‘myInstitutionOnto’).
Once you have
created this extension, if you think the components added can be of general
interest and would like the EGAD to
integrate into RiC-O some or all of the additions made, let us know what you
have done (send an email to this list, or to us directly, or else create an issue on
GitHub).
Or else, before creating any extension, just create an issue on GitHub
and share your ideas or proposals with the RiC-O EGAD team. As we have already
done in other cases, we will then be able to discuss them with you and see what
we can do (I mean, if we agree on adding what you suggest directly to RiC-O). Note
that such discussions and work might take us a while - we don't plan an
intensive phase of RiC-O updates this summer, except perhaps to start working
on new vocabularies to populate some RiC-O classes (rico:ActivityType for
example).
Hope this helps you!
Best regards,
Florence
---
Florence Clavaud
Executive member of ICA/EGAD ; lead of RiC-O development team
Conservatrice générale du patrimoine | General curator
Responsable du Lab des Archives nationales de France| head of the Lab, Archives nationales de France
Kind regards,Arian