Check-in and access this session from the IGF Schedule.

IGF 2019 WS #222
Multistakeholder Governance for Semantic Interoperability

    Subtheme

    Organizer 1: Denny Vrandecic, Google
    Organizer 2: Lydia Pintscher, Wikimedia Deutschland
    Organizer 3: Max Senges, Google

    Speaker 1: Vint Cerf, Private Sector, Western European and Others Group (WEOG)
    Speaker 2: Lydia Pintscher, Civil Society, Western European and Others Group (WEOG)
    Speaker 3: Dan Brickley, Private Sector, Western European and Others Group (WEOG)

    Moderator

    Max Senges, Private Sector, Western European and Others Group (WEOG)

    Online Moderator

    Denny Vrandecic, Private Sector, Western European and Others Group (WEOG)

    Rapporteur

    Max Senges, Private Sector, Western European and Others Group (WEOG)

    Format

    Round Table - U-shape - 60 Min

    Policy Question(s)

    Data Governance: how can we create an appropriate framework that will allow us to expediently and effectively increase Semantic Interoperability between various diverse stakeholders?
    How can we build this framework so that the parts that require sustained maintenance and funding to remain relevant are, in fact, sustainably funded and supported?

    SDGs

    GOAL 9: Industry, Innovation and Infrastructure
    GOAL 11: Sustainable Cities and Communities
    GOAL 12: Responsible Production and Consumption
    GOAL 16: Peace, Justice and Strong Institutions
    GOAL 17: Partnerships for the Goals

    Description: Participants will deliberate about a shared vision and meaningful cooperation as well as the adequate governance model for this effort. Namely the following questions, that need to be solved in order to achieve Semantic Interoperability, will be addressed:
    What are viable elements of a joint vision for the syntax and semantics of the basic data model? That includes the validation and easy reuse of data sets, a catalog of entities, shared vocabularies i.e. the ontology of the data set.
    How do we collaborate to identify and include entities outside of the catalog?

    For the workshop we will prepare a straw-man solution - using Wikidata as the common catalog of entities and schema.org for the common vocabulary governance model -, which we will distribute in advance and invite feedback and discuss various aspects and welcome alternative ideas.


    Expected Outcomes: Agree on a joint vision and either (ideally) agree on the strawman proposal in order to move forward, or identify the requirements that are lacking, so that the research and development communities can be tasked with finding solutions that fulfill these identified requirements.

    Bootstrap a group of stakeholders interested to cooperate to pursue the vision across the existing efforts. Namely the idea is not to create a new entity but rather to allow each group to follow its interest and mandate while staying connected as a community not last to address topics of inclusion and open standardization.

    We will publish the strawman proposal before hand. We will also provide an informal tutorial on the strawman proposal in order to ensure a productive discussion.

    Relevance to Theme: To agree on how we describe the (digital) world might be one of the greatest human endeavours. In some ways aiming for semantic interoperability can be compared to re-unite Babylon - as the aim is to create a digital lingua franca or a Rosetta Stone that would lay the foundation for digital services to interoperate on the data layer. In the perspective of internet architecture, it has been argued that semantic interoperability could be seen as the second necessary common element next to TCP/IP.

    Semantic Interoperability requires that we agree apriori on the syntax and semantics of the data formats that we use. But in order to ensure the inclusion of all stakeholders, we must ensure that we have formats that have the right extension points and allow for a diverse community to use these extension points flexibly. The extension points must be in the entity catalogs and the vocabulary used on top of the common syntactic layer.

    We want to deliberate about the right governance models and ecosystem design to ensure that semantic interoperability is pragmatically simple enough to use and participate in, but at the same time stable enough to be effective.


    Relevance to Internet Governance: In order to ensure Semantic Interoperability, we need to agree on sufficient minimal standards for the syntax and semantics of data formats, and to agree on the governance of the extensible vocabularies authority catalogs.

    One of the most valuable but also expensive tasks when integrating datasets is to ensure the semantic interoperability of the data: the reconciliation of the relevant entities and the vocabulary being used. Traditionally, a lot of interoperability work has been focused on syntactic interoperability - making sure that the formats one tool writes can be read by another tool. This is absolutely crucial, but insufficient. Semantic interoperability is given when we know that the entities referred to in one resource can be safely mapped to entities referred to in another resource.

    This is in general a very difficult “technical” problem. In this workshop we are bringing together stakeholders from industry, academia, standards organisations and civil society in order to bootstrap work on a solution that is ideally decentralized in terms of institutional ownership, sustainably maintained for the long term, focused on efficiency but is also considerate of human rights and diversity.

    Online Participation

    We will make an open call to the Wikimedia communities and other interested communities, inviting them to join the participation tools. Denny Vrandecic will be working on the online participation tools.

    Proposed Additional Tools: We plan to host a copy of the strawman proposal in an open fashion on the Wikimedia wikis, so that anyone can participate there, and to gain further interest and input.