Re: Multiligual Credentials

Dear All,

For those of you who couldn't make it to our last VC for Education Task Force meeting on the 2nd of December, and those who are interested in a succinct recap, below I am sending some references shared in the call where we were talking about how we build multilingual verifiable credentials in Europe. If you'd like to consult the whole recording and transcript, you can find them at https://w3c-ccg.github.io/meetings/2024-12-02-vc-education/.

As we speak many different languages in Europe, and some countries have more than one official languages (e.g. English, French and German in Luxembourg), the European Learning Model<https://europa.eu/europass/elm-browser/index.html> (ELM) is multilingual by design - in December 2024, the standard supports 29 languages, and two new languages are to be added in 2025.
In addition to the model being multilingual, for a number of concept type credential properties, such as e.g. the designation of a qualifications framework level, subject areas<https://op.europa.eu/en/web/eu-vocabularies/concept-scheme/-/resource?uri=http://data.europa.eu/snb/isced-f/25831c2> (also demoed briefly during the call), or the type of accreditation, activity, assessment, etc., the ELM's European Digital Credentials<https://europa.eu/europass/elm-browser/homepage/edc-generic-no-cv_en.html> application profile uses multilingual controlled vocabularies, that are published on the official EU Vocabularies portal of the European Union. The Common European Skills Data Space<https://op.europa.eu/en/web/eu-vocabularies/europass>, that the ELM Team manages, currently hosts 34 controlled lists<https://op.europa.eu/en/web/eu-vocabularies/europasstables>. Furthermore, some ELM properties are expecting values from other external taxonomies and authority tables, such as the ESCO skills<https://esco.ec.europa.eu/en/classification/skill_main> and occupations<https://esco.ec.europa.eu/en/classification/occupation_main> taxonomies, the Administrative territorial unit (ATU)<https://op.europa.eu/en/web/eu-vocabularies/concept-scheme/-/resource?uri=http://publications.europa.eu/resource/authority/atu> authority table, the Countries and territories<https://op.europa.eu/en/web/eu-vocabularies/concept-scheme/-/resource?uri=http://publications.europa.eu/resource/authority/country> and Language<https://op.europa.eu/en/web/eu-vocabularies/concept-scheme/-/resource?uri=http://publications.europa.eu/resource/authority/language> controlled vocabularies. As we mentioned in the call last Monday, the use of multilingual lists, as opposed to automatic translation, is very important, because some terms may be sensitive or domain specific, and for this reason the ELM lists' concepts have all been translated by professionals. Other reasons why we use controlled vocabularies in EDC metadata, is to support data transparency, credential recognition and portability, and to allow cross referencing without manual intervention (e.g. ESCO occupations are linked to essential and optional knowledge, skill and competence concepts, and National Qualifications Framework<https://op.europa.eu/en/web/eu-vocabularies/concept-scheme/-/resource?uri=http://data.europa.eu/snb/qdr/25831c2> levels are mapped to European Qualifications Framework<https://op.europa.eu/en/web/eu-vocabularies/concept-scheme/-/resource?uri=http://data.europa.eu/snb/eqf/25831c2> levels).

If you want to see how multilingual credential content is displayed in the EDC infrastructure, please find attached a sample micro-credential that you can upload to the EDC Viewer<https://europa.eu/europass/digital-credentials/viewer/#/home>. There you can choose any of the 29 browsing languages to contextualise the content (click on the EN bubble in the top right corner of the page highlighted with a red square on the illustration below), and - independently from your chosen browsing language - choose any of the available credential content languages from the drop-down list highlighted with a yellow square. Note, that an EDC's content always must have 1 primary language (that isn't always English) and may have up to 28 additional languages, as provided by the credential issuer. The attached credential sample is a relatively simple micro-credential. In case you are interested other EDC examples, the ELM Team has a (growing) list of samples in this repository<https://code.europa.eu/qualifications-courses-and-credentials/ELM-support/-/tree/main/credential%20examples/JSON-LD%20Examples%20(ELM%20v3)?ref_type=heads>.


[cid:6b089dd3-93ea-4b82-b7d2-8b6bdf9b8075]
Would you have any reflections or observations on any of the points above, please share, let's keep the conversation going.

Best wishes,

Ildiko


Ildiko Mazar

Sector Consulting Manager

European Digital Credentials for Learning Implementation Support

NTT DATA
Rue de Spa 8, 1000 Brussels, Belgium
Learn more at nttdata.com<https://urldefense.com/v3/__https:/www.nttdata.com/global/en/__;!!DOxrgLBm!SSn61vBA3b9R8E1QbsOrqGN-KqUb93Wb7c6tOhqNMCZoHVO9lPxtTID8cooVCFOQ2dq8xXCU$>

[cid:012e98e8-930c-4662-ba25-409df6f03d76]



________________________________
From: Kerri Lemoie <klemoie@mit.edu>
Sent: 22 November 2024 19:09
To: public-vc-edu@w3.org <public-vc-edu@w3.org>; public-credentials@w3.org <public-credentials@w3.org>
Subject: Multiligual Credentials

NTT DATA Security Awareness - This is an incoming mail from an EXTERNAL SENDER (public-vc-edu-request+bounce-ildiko.mazar=nttdata.com@listhub.w3.org). Please verify sender before you open attachments or access links.

________________________________

Hello All,



At the November 18, VC-EDU call, we discussed approaches to multilingual credentials (transcription and audio available through this tool: https://w3c-ccg.github.io/meetings/scribe-tool/<https://urldefense.com/v3/__https://w3c-ccg.github.io/meetings/scribe-tool/__;!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0JnGEzKZ$>).



I believe this is an important topic especially for education related credentials.



I’d like to kick-off a discussion on this thread and consider a follow-up discussion at VC-EDU again and/or CCG soon. Together we can consider pragmatic approaches that will help to make the credentials as useable and understandable as possible.



Primarily we discussed two ways that I’m aware of that we could issue credentials in multiple languages.



W3C VCDM approach: https://www.w3.org/TR/vc-data-model-2.0/#example-example-dual-language-credential<https://urldefense.com/v3/__https://www.w3.org/TR/vc-data-model-2.0/*example-example-dual-language-credential__;Iw!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0KI9YXz9$>



The angle of this approach is to provide translations within the properties themselves.



1EdTech Open Badges 3.0 & Comprehensive Learner Record V2: https://gist.github.com/ottonomy/12bd1e1f5510d3f49d966b1991f1e6c8<https://urldefense.com/v3/__https://gist.github.com/ottonomy/12bd1e1f5510d3f49d966b1991f1e6c8__;!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0Dn_YxA6$> (draft gist provided by Nate Otto – thanks)



This approach specifies creating related achievements that could be in different languages using:

  *   related (https://www.imsglobal.org/spec/ob/v3p0#org.1edtech.ob.v3p0.related.class<https://urldefense.com/v3/__https://www.imsglobal.org/spec/ob/v3p0*org.1edtech.ob.v3p0.related.class__;Iw!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0MUjO1Pq$>)
  *   with a languageCode: (https://www.imsglobal.org/spec/ob/v3p0#org.1edtech.ob.v3p0.derived.languagecode.class<https://urldefense.com/v3/__https://www.imsglobal.org/spec/ob/v3p0*org.1edtech.ob.v3p0.derived.languagecode.class__;Iw!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0F-EAY0L$>)



It’s worth noting this approach has been used by the 1EdTech specs prior to the more current versions that are aligned to VCs and that they could be open to changing this in a subsequent version.



Also worth noting that the current Open Badges 3.0 and CLR v2 (presumably since they are very similar) schemas will not allow for the VCDM method because the properties where you would likely apply this are expected to be strings versus arrays or objects. So, any software verifying against the schemas will return an error if say, both the VCDM and the 1EdTech angles to multilingual credentials are used simultaneously.

Would love to hear from you about it.



Best,



K.





---

Kerri Lemoie, PhD
Director – MIT, Digital Credentials Consortium
https://digitalcredentials.mit.edu<https://urldefense.com/v3/__https://digitalcredentials.mit.edu__;!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0LX0PuKR$>
she/her

The DCC is advancing the use and understanding of portable, verifiable digital credentials in higher education through open source technology development and leadership, research, and advocacy.



Join the DCC mailing list<https://urldefense.com/v3/__https://mit.us6.list-manage.com/subscribe?u=ad81d725159c1f322a0c54837&id=3621913fe4__;!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0FVQeeSJ$> and follow us on LinkedIn<https://urldefense.com/v3/__https://www.linkedin.com/company/dccconsortium__;!!EJ3n55FBLexp1rhr!6NQmxSGhpQ6GHhqRVNUQbIk43XwFg9z4AALj_3fnAtZ9AYG0rjmjuBYPO_dXIxVtknqZ3B6REMVOHW_O0CNSSKeU$> for news & updates.