Wikidata:Development plan/archive2022/status updates

Wikidata as a platform

Empower the community to increase data quality

We must ensure that our socio-technical system helps editors increase the quality of Wikidata’s existing data and contribute new high-quality data.

Feedback loops with data re-users

Objective:

We want to build up feedback loops with data re-users to get them actively involved in improving the data on Wikidata.

Description:

We are developing the Mismatch Finder as a base for setting up these feedback loops.

Upcoming Activities:

• Release Mismatch Finder (Q1)
• Work with data re-users and researchers to get mismatches to load into the Mismatch Finder (Q1, Q2)
• Make adjustments to the Mismatch Finder based on feedback we are receiving on the first version (Q2, Q3)
• Start work on v2 of Mismatch Finder (Q4)

Outcomes:

• We released the Mismatch Finder and made several large improvements to the tool based on feedback from editors and people providing mismatches. Over 30.000 mismatches have been made available in the tool so far. We will still need to work more with partners to receive data quality improvements through the tool regularly and with editors to make good use of the available mismatches for review.

Entity Schema build-out and integration

Objective:

We want to better integrate the existing Entity Schemas so that they become more central to data maintenance processes and can codify modeling decisions.

Description:

We have Entity Schemas on Wikidata that are not well integrated to existing workflows and community processes. We need to decide what the next steps are.

Upcoming Activities:

• Run a team workshop to get everyone up-to-speed with Entity Schemas (Q1)
• Decide on a technical way forward for Entity Schema development (Q1, Q2)
• Plan the development of version 2 of EntitySchemas (Q3)

Outcomes:

• We have planned the most significant improvements that are part of version 2 of Entity Schemas. We are still exploring various technical options for making them possible and will continue the work on this in 2023.

Data quality and data utility processes

Objective:

We want to better understand the different processes and tools that help with increasing/maintaining data quality and utility to improve them together where needed.

Description:

There are a lot of tools and processes around data quality and utility in Wikidata. We need to thoroughly understand them and have conversations about how to improve them and where improvements are most necessary.

Upcoming Activities:

• Collect information and understand the current landscape (Q2)
• Evaluate the input from the Data Quality Days (Q3)

Outcomes:

• We ran the Data Quality Days with a lot of very useful discussions. The input from that will help us plan further data quality efforts.
• We have made several improvements to existing systems like the constraints violations.
• We have prepared a survey for data re-users to better understand the worst ontology issues they are facing. This will be run in Q1 2023 and then brought back for discussion with editors for the next steps.

Data Quality Days

Objective:

We want to continue to provide a platform for the community to have discussions that are focusing on improving the usefulness of the data in Wikidata. We will run a follow-up event to the 2021 Data Quality days.

Description:

The sessions will take place online from July 8-10, with a flexible program allowing people to propose and schedule sessions on topics related to making Wikidata’s data more useful and maintaining its high quality. More information about the topic of the event, ways to contribute to the sessions and the date will be communicated in Q2.

Upcoming activities:

• Communicate about the event (Q2)
• Organise and run the event (Q3)
• Share a summary of the conversations (Q3)

Outcomes:

• We ran the Data Quality Days with a lot of very useful discussions. The input from that will help us plan further data quality efforts.

Strengthen underrepresented languages

More people need access to knowledge and technology presented in their own language, and content in that language should be accessible to all. Language data is a fundamental building block in reaching that goal.

Lexicographical Data UI improvements

Objective:

We will make it easier for editors to create new Lexemes.

Description:

We will improve Special:NewLexeme.

Upcoming Activities:

• Development of a new version of Special:NewLexeme (Q1, Q2)
• Roll-out of the Special:NewLexeme page and addressing feedback from testing (Q3)
• Set up interviews with editors for testing and to collect additional feedback (Q3);

Outcomes:

• We have released a new Special:NewLexeme page and made further improvements to it based on editor feedback.
• We have made various smaller improvements to the Lexeme page as well as the Query Service intended to make working with lexicographical data easier.

Setting up a Wikidata Software Collaboration (I)

Objective:

The decision around who will be the partners in the shared software collaboration with another movement partner has been made, communicated to the movement and questions have been answered.

Description:

Wikimedia Deutschland wants to create a new partnership with another movement partner to start sharing responsibilities and resources for Wikidata software development.

Upcoming Activities:

• Discussing possible partnership with interested groups (Q1)
• Setting up the partnership and the necessary infrastructure and paperwork for this (Q2)

Outcomes:

• Two partners from the movement have entered a partnership with Wikimedia Deutschland: Wikimedia Indonesia and the Igbo Wikimedians User Group.

Setting up a Wikidata Software Collaboration (II)

Objective:

A movement partner has committed to the planned Software Collab and all contractual/ financial documents.

Description:

We are setting up the necessary infrastructure for a software team to start their work in another movement organization: Hiring staff, communication channels WMDE and movement partner, definition of roles and responsibilities of the partners.

Upcoming Activities:

• Update the community about which movement partner is selected for the collaboration (Q2)
• Setting up the infrastructure for comprehensive onboarding to Wikidata software development (Q2)
• Staff at Wikimedia Indonesia are onboarded and have the necessary information and resources to start scoping their work on Wikidata/ Lexeme software (Q3)
• The Igbo Wikimedians User Group have the necessary information and resources to implement Wikidata:Wiki Mentor Africa (Q3)

Outcomes:

• Wikimedia Indonesia is working on building up a software team on their own and have completed their user research to understand their community's needs better when it comes to Wikidata and especially Lexemes.
• The Igbo Wikimedians User Group are successfully implementing their mentoring program for communities in Africa to enhance the capacity for Wikidata tool development in their communities.

Increase re-use for increased impact

We want to make sure that anyone can use the data in Wikidata to make the world a better place. While everyone can re-use our data, we give priority to organizations and projects that align with our values and have a high impact.

REST API

Objective:

We want to improve our APIs to make it easier for programmers to access our data.

Description:

We will implement the REST API proposal.

Upcoming Activities:

• Start implementation of the REST API (Q1)
• Continue implementation of the first version of the Wikibase REST API (Q2)
• Deploy a first version for testing and feedback during development to beta Wikidata (Q3)

Outcomes:

• We have developed the first version of the Wikibase REST API and made it available for testing. We have received a lot of useful feedback that helped improve the first version. The release of the new API to Wikidata will happen in Q1 2023.

Data Reuse Days

Objective:

Following the “X Days Event” format (after 30 Lexicodays and Data Quality Days in 2021), the Data Reuse Days 2022 is a series of online events taking place on March 14-24 and focusing on applications using Wikidata’s data.

Description:

The sessions will take place online from March 14 to 24, with a flexible program allowing people to propose and schedule sessions on the topic of using Wikidata’s data.

Upcoming activities:

• Run the event (Q1)

Outcomes:

• We have run a multi-day event with talks, demo sessions and discussions to help bring data reusers closer to the Wikidata community. The event was very successful and helped everyone better understand what reusers need from Wikidata and which problems they are facing. It also highlighted a lot of really great ways Wikidata’s data is used.

Improved documentation for data re-users

Objective:

We want to help more people to be able to build applications and services on top of Wikidata.

Description:

We want to improve the documentation for data re-users so they have an easier time building applications on top of Wikidata’s data.

Upcoming activities:

• Improve documentation for deciding which access method to choose for accessing our data (Query Service, dumps, action API, …) (Q1)

Outcomes:

• A new data access page has been published that better explains the different ways to work with Wikidata’s data and when each of them is appropriate. The same content was additionally presented at the Data Reuse Days.

Wikibase Ecosystem

Empower knowledge curators to share their data

Increase the number and diversity of Wikibases that can eventually be connected to the LOD web.

Wikibase.cloud Launch (Initial)

Wikibase.cloud is a Wikibase as a Service platform based on open-source WBstack code, but owned and maintained by Wikimedia Deutschland.

Objective:

Users can onboard into the Wikibase ecosystem without a complex installation and set-up process.

Description:

We will prepare a "minimum launchable platform" consisting of technical infrastructure, required processes, and plans for maintenance so that we can begin onboarding new users.

Upcoming Activities:

• Prepare and launch a closed beta version of Wikibase.cloud, including a managed waiting list. The waitlist will be considered after WBStack migration has completed (Q1)
• Wikibase.dev staging area is online (Q1)
• Wikibase.cloud initial infrastructure deployment (Q1, Q2)
• Actually launch the closed beta version of Wikibase.cloud. If you want to apply to get closed beta access once the migration of the first users is over, please fill out this form. (Q2)
• Upgrade Wikibase Suite and Wikibase.Cloud so they are up to date with MediaWiki releases (Q3)
• Release security updates for Wikibase.Cloud (Q3)

Outcomes:

• Wikibase.cloud is live!
• In addition to having the platform up and running, we’ve experienced very few hiccups and no considerable downtime
• We started onboarding new users post-migration in September, and have been continuously welcoming people who apply for early access ever since

WBstack migration

Objective:

Users can onboard into the Wikibase ecosystem without a complex installation and set-up process.

Description:

Devise a migration path for WBstack users to Wikibase.cloud, offer this option to user base, and move them to the cloud platform so that WBstack.com can be sunset.

Upcoming Activities:

• Current users of WBStack receive an email with details on how you can opt-in to a migration to wikibase.cloud in March (Q1)
• Migrate all WBStack users who opted in to being migrated to Wikibase.cloud (Q2)

Outcomes:

• The migration of the Wikibases on WBStack that wanted to migrate to Wikibase Cloud was completed in June and a great success!

Wikibase.cloud Launch

Objective:

Users can onboard into the Wikibase ecosystem without a complex installation and set-up process.

Description:

Ensure technical reliability and data-driven insights of Wikibase.cloud to support further growth.

Upcoming Activities:

• Onboard 50 waiting list users (Q3)
• We have an inventory of all technical issues blocking essential user experience, and solve them as we go (Q3)
• We have monitoring in place for essential metrics related to the platform performance and user experience (Q3)
• Investigated the provision of a suitable search experience and drafted a solution (Q3, Q4)
• Conducted user research to understand Wikibase.cloud target users and problems we want to solve going forward, to establish a roadmap (Q3, Q4)
• Created a logo for Wikibase Cloud (Q4)

Outcomes:

• We have welcomed close to 150 new Wikibases since starting actively sending out invitations in September, bringing us to over 400 Wikibases hosted on Cloud
• We have a first version of dashboards that help us closely monitor the status of the platform and actively alerts us when there’s trouble or when we need to scale some resources. It also allows us to see high-level activity to monitor how active our user base is, for example in the amount of Wikibases, editors and such
• We wrapped up a survey in late 2022 that gave us very valuable insights into how our users are experiencing Wikibase Cloud and where we can improve, which is the focus for our plan in 2023
• The survey also showed that people are happy with the ease of setting up a Wikibase, especially compared to the self-install alternative that was previously the only option
• We announced our logo as a New Year’s present
• Search is our top priority for early 2023 to solve

Establish a process for expanding into new regions and onboard new non-EU/NA partners

Objective:

A truly global LOD web requires truly global connections and increased access to marginalized knowledge. This requires partnerships to move beyond North America and Europe.

Description:

We have already begun pushing into India and are beginning to make connections there.

Outcomes:

• Formalise connections with the community and leveraging those into GLAM connections (Q1, Q2)
• We established connections with two principal players within the Indian community - Bohdisattwa (West Bengal Wikimedians User Group) and Subodh (CIS A2K) - and with some Indian GLAMs. However, we struggled to connect with the Indian community at large.
• We established good relationships with the WMF representatives for India.
• We spread our connections beyond India and helped several institutions outside of NA/EU/India such as the Kenya National Library Services to establish and maintain their own Wikibases

Wikibase Market Research

Objective:

Better understand which organisations want to use Wikibase in the future and for what

Description:

We want to better understand where there is potential for the Wikibase Ecosystem to grow and how, both in existing areas (GLAM, sciences, and digital humanities) as well as new areas in which there are no established Wikibase projects yet. This foundational research will focus both on researching alternatives to Wikibase and understanding how organisations adopt software.

Upcoming Activities:

• Work with an agency to investigate new use cases for Wikibase and existing alternative solutions to Wikibase (Q1)
• Share report with the community (Q2)

Outcome:

• Futur2 Market Research
• GLAM was highlighted as not being the future of Wikibase, therefore Data Partnerships opened up to more knowledge fields.
• The report was shared with the community here

Ecosystem Enablement

Enable an ecosystem of extensions as well as tools and custom interfaces based on WB APIs to emerge around Wikibase, extending the functionality of the software for more use cases. Ecosystem Enablement Conversations

Objective:

External developers have a clearer picture of which possibilities for tools and extensions are in line with the development of the Wikibase software.

Description:

We will establish a successful exchange of information between the Wikibase team and external developers and other stakeholders building extensions and tools so that we understand their plans, challenges, and requirements when extending the functionality of Wikibase.

Upcoming Activities:

• First steps toward “A successful exchange of information between the Wikibase team and external developers and other stakeholders building extensions and tools” is established. E.g. workshops or other systems for facilitating this exchange (Q1)
• At least one workshop has happened (Q2)
• Generate workshop documentation for developers (Q3)

Outcome:

• We ran two workshops. One on working with mwcli and the other on installing extensions to Wikibase
• We created documentation for developers which can be found here and here
• We continue to work on documentation for workshop creators, so that others apart from us can run similar workshops in future

Connect Data across technological & institutional barriers

Ensure Wikibases can connect more deeply with each other and Wikidata to form an LOD web. User testing of Federated Properties in combination with local properties

Objective:

Wikibase instances outside of Wikimedia use Wikidata to augment and enrich their own data.

Description:

Federated Properties v2 is an experimental version of federation that enables a Wikibase to use Wikidata's properties in combination with a local data model. We will facilitate testing of the feature by institutions and projects to gather valuable feedback, with the goal of later implementing improvements to the user interface and experience.

Upcoming Activities:

• We’ve set up an environment on Wikibase.dev where the version of Federated Properties 2 that existed at the end of our 2021 hike can be tested by institutions being recruited from the WBSG (Q1)
• This feature is tested by at least 3 projects (Q1, Q2)

Outcome:

• We tested federated properties with 4 representatives of institutions. All participants emphasized the need to differentiate between local and federated properties.