W3C

Data Privacy Vocabularies and Controls Community Group Teleconference

12 November 2019

Attendees

Present
Bert
Regrets
-
Chair
Axel
Scribe
Axel

Meeting minutes

<harsh> Hello, having a few issues with calling.

<harsh> Okay, I'll try going to a different wifi location

<axelpolleres> again late, sigh

<axelpolleres> dialing

<axelpolleres> anyone still here? ☺

<Bert> Hi Axel, just Rudy and me

<Bert> Harsh was trying to find wifi

<axelpolleres> Attendees: Harsh, Ber, Rudy, Axel

<axelpolleres> 2 main agenda items today... IDS presentation/discussion, mail by rudy and discussion of additonal terms in the vocabulary.

IDS

<axelpolleres> christoph lange presenting.

<axelpolleres> IDS - Industrial Data Space

<axelpolleres> potential synergies between IDS information model and DPV

<axelpolleres> IDS information model is open, so no IP issues.

<axelpolleres> http://ids.semantic-interoperability.org/ is not the latest version of the information model.

<clange> IndustrialDataSpace/InformationModel

<harsh> clange: Latest version is at IndustrialDataSpace/InformationModel

<axelpolleres> data and usage control is important

<axelpolleres> for IDS

<clange> https://www.internationaldataspaces.org/info-package/

<clange> https://www.internationaldataspaces.org/ressource-hub/publications-ids/

<clange> Reference Architecture Model or RAM

<clange> (showing some improved graphics from that)

<axelpolleres> policy compliance checking in IDS still work in progress, policies ODRL-based.

<harsh> axelpolleres: for the vocabulary used, how is it modeled? Does it have a hierarchy/taxonomy?

<harsh> clange: not at this point

<harsh> clange: this is where DPV could be useful

<axelpolleres> DPV complementary... IDS doesn't have concrete classes of data providers/consumers defined, nor putpose hierarchies.

<axelpolleres> IDS mainly B2B scenarios, but also personal data could be exchanged, where DPV would be important to trace these exchanges, which need to be documented according to GDPR, for instance.

<axelpolleres> IDS could look into re-use and extend DPV

<axelpolleres> Christoph: DPV highly useful for that

<axelpolleres> ... IDS working on an app store on their data market, DPV useful for metadata on the app store.

<axelpolleres> ... also for describing datasets.

<axelpolleres> also Datacube, void useful for that.

<axelpolleres> ... DPV relevant for personal/privacy related data.

<axelpolleres> ... in the next few months more use cases for IDS information model.

<clange> further public information about the IDS Information Model: https://www.internationaldataspaces.org/wp-content/uploads/2019/07/IDS-Information-Model.pdf, https://www.internationaldataspaces.org/wp-content/uploads/2019/07/IDS-Policy-Language.pdf – both slightly outdated, but combine this with the https://github.com/IndustrialDataSpace/InformationModel/blob/master/CHANGELOG.md

<clange> … and then you are up to date wrt. the current version 2.0

<axelpolleres> ... can imagine a lot of useful ways to use DPV, do not expect disruptive new.

<clange> wrt. the specific use of ODRL, we basically extended ODRL's supply of Left Operands: https://github.com/IndustrialDataSpace/InformationModel/search?q=leftoperand&unscoped_q=leftoperand

<axelpolleres> Christoph trying to join the group, IDS association has a lot of industry (more than 100 companies).

<axelpolleres> no standardisation body behind IDS, but reusing a lot of existing standards.

rudy's suggestion to additional terms to the vocabulary.

<axelpolleres> Link to the mail?

<harsh> https://lists.w3.org/Archives/Public/public-dpvcg/2019Nov/0002.html

<harsh> further discussion in https://lists.w3.org/Archives/Public/public-dpvcg/2019Nov/0003.html

<clange> Re. standardisation of IDS: the Reference Architecture Model is a high-level standard-like document; for the 100+ member organizations of the IDSA, technical guides with more implementation-related guidelines are currently being elaborated (and will at some point also go public).

<harsh> Suggested terms were regarding TV Viewing Behaviour (Personal Data Category) and Creating Event Recommendations (Purpose)

<clange> The information model itself basically reuses and extends a lot of W3C standards, including DCAT, ODRL and SKOS.

<axelpolleres> harsh: personalization and recommendation should be different purposes.

<harsh> Currently, we have a type of persaonlised recommendations under ServicePersonalisation

<axelpolleres> PROPOSED: personalization and recommendation should be different purposes.

<axelpolleres> +1

<harsh> +1

<clange> Some results of the IDS-related projects are themselves also being fed into "proper" standardization processes, e.g., https://www.internationaldataspaces.org/din-spec-27070-close-to-be-published/ (DIN SPEC 27070 "A standard for secure gateways for data and services")

<axelpolleres> axel: recommendation could be extended further, but maybe not extend the scope now.

<Bert> +1

<harsh> dpv:CreateProductRecommendations is the current concept

<harsh> dpv:CreateProductRecommendations ⟶ ServicePersonalisation ⟶ Purpose

<axelpolleres> dpv:servicePersonalPersonalization is currently superclass of creat product recommendation.

<axelpolleres> rather: personalizedrecommendation subclass of botrh.

ACTION: HArsh will make a formal proposal for an extension/modification in this direction.

<trackbot> Created ACTION-135 - Will make a formal proposal for an extension/modification in this direction. [on Harshvardhan Pandit - due 2019-11-19].

<axelpolleres> a change would break some dependency, we need to document that change well example.

<axelpolleres> HArsh will organise a sync call with Rudy to make the proposal more conrete by next time.

<axelpolleres> if anybody wants to join, then let them know.

<harsh> I (We) will send details if anyone wants to join on the mailing list

<axelpolleres> Please inform us on the mailinglist about the outcome... ideallty in a mail we can vote on next time.

<clange> Regarding the potential significance of something (e.g. the DPV) being recommended in the official examples about the IDS Information Model, note https://www.internationaldataspaces.org/the-association/#mitglieder (100+ members)

<axelpolleres> next call 26th, Axel at risk, Bert can chair.

<axelpolleres> adjourned.

<clange> The slides I presented by screen-sharing are a variant of what https://www.internationaldataspaces.org/wp-content/uploads/2019/03/IDS-Reference-Architecture-Model-3.0.pdf calls the "system layer"

<clange> (section 3.5)

<axelpolleres> Axel: if I still find an open issue/action wrt. IDS liaison, I would close it for now and hope for Christoph to be in touch, following up the call today.

<axelpolleres> i.e.: see/read/hear you then @clange 😉︎

<clange> great, thanks, indeed let's keep in touch

<axelpolleres> minuts are out there: https://www.w3.org/2019/11/12-dpvcg-minutes.html

<axelpolleres> BTW, I reaslize we forgot tyo approve the last meetings minutes:

ACTION: Axel to remind Bert to approve both last meetings' minutes next time (26 November)

<trackbot> Created ACTION-136 - Remind bert to approve both last meetings' minutes next time (26 november) [on Axel Polleres - due 2019-11-19].

Summary of action items

  1. HArsh will make a formal proposal for an extension/modification in this direction.
  2. Axel to remind Bert to approve both last meetings' minutes next time (26 November)
Minutes manually created (not a transcript), formatted by scribe.perl version 217 (Fri Apr 7 17:23:01 2023 UTC).