2019:Hackathon/GLAM focus area

Tracked in Phabricator
Task T215409

The Wikimania 2019 Hackathon in Stockholm, August 14–15, 2019, has GLAM (Galleries, Libraries, Archives, Museums) as one of its focus areas.

Wikimedia developers, and developers working for cultural institutions, are very welcome to join this focus area. Let's sit together in the hacking space, and let's inspire each other!

Support and general topics of interest

You are free to work on the projects you are interested in. For two specific topic areas, additional support will be available, with staff from Wikimedia Sverige and the Wikimedia Foundation present.

FindingGLAMs

The FindingGLAMs project is being run by Wikimedia Sverige and aims to both identify GLAM institutions around the world and try to identify issues preventing GLAM institutions from contributing with materials. Through the analysis being done in this project, several areas of work have started to crystallise as in need of more attention. Wikimedia Sverige can give some support for participants wishing to work with any of these areas; they are indicated with (FindingGLAMs support) below.

Structured Data on Wikimedia Commons

In 2019, Wikimedia Commons is undergoing a major update: it is now possible to describe files on Wikimedia Commons with multilingual, structured, machine-readable data from Wikidata. The Wikimedia Foundation can give some support for participants wishing to work in any of these topics; they are indicated with (SDC support) below.

Project proposals and ideas

The list below is not exhaustive. Please edit this page and add other topics!

Also make sure to add your tasks to the hackathon's Phabricator board: https://phabricator.wikimedia.org/project/board/3922/

Sharing Audio files and Sheet Music Globally (FindingGLAMs support)

Many GLAMs hold collections of e.g. sheet music and performance recordings. While the Wikimedia platforms support digitised musical scores, the process of actually digitising a scanned sheet is quite involved today.

Specific steps for facilitating this could involve:

Making OMR available through Wikisource (similar to OCR today)
Addressing issues with displaying sheet music in Wikisource/Commons
Investigating how to best tie together a scanned musical score with its digitized counterpart and a recorded performance of the same work.
Creating comprehensive documentation that takes up all the steps involved in digitizing a scanned musical score

Import support for lexical datasets (FindingGLAMs support)

With the WikibaseLexeme extension largely stabilized, time has come to implement support for it in the major import tools (for example Pywikibot).

GLAM material in this area comes both in the shape of highly structured data and as scanned lexicons. Either could be used to import or source lexicographical data in Wikidata but both are in need of tooling to facilitate the process. For the latter type of material it would be of interest to see how the fact that the source material is digitally available on one of the sister platforms, e.g. transcribed at Wikisource, could best be leveraged.

Sharing documents used for sourcing (FindingGLAMs support)

Uploading archive materials to Wikisource (via Wikimedia Commons) allows it to be transcribed and used as sources on Wikipedia and Wikidata. This is especially important for languages in which most important sources are not yet available digitally.

Tooling/documentation which facilitates the use of such material for reference purposes is however sorely missing.

Engaging people in addressing problematic data (FindingGLAMs support)

Much of the uploaded GLAM materials comes with original data or descriptions which are marked as potentially being problematic or even offensive due to the time and context in which it was originally created. While the original description is valuable in providing the context in which the material was created or collected the Wikimedia platforms provide a good opportunity to review historical descriptions and update the material so that it better relates in our time.

Tools making it easier to bring material needing review to the surface and facilitating addressing such, e.g. through micro contributions, would be of great value.

Hacking with Structured Data on Wikimedia Commons (SDC support)

Tools for uploading files to Wikimedia Commons with structured metadata (FindingGLAMs support) (SDC support)

With Structured Data on Commons now being available the time has come to add support for it in the major upload tools. The new development also brings to the surface the need for tooling and documentation to help support the preparation of LAM metadata and structuring/mapping it to Wikimedia Commons/Wikidata metadata.

Improve the process of doing bot-driven uploads to Wikimedia Commons
Upgrades to Pywikibot to support structured data
Integrate structured data uploading and editing to Wikimedia Commons in OpenRefine
Integrate structured data uploading in Pattypan
Uploading via the Monumental tool
Help with bot-driven uploads for GLAM pilot projects of Structured Data on Commons
...

Tools for editing Wikimedia Commons with structured metadata (SDC support)

Help improve the ISA microcontributions and tagging tool for Wikimedia Commons
...

Improving data re-use between Wikidata, Structured Data on Commons and other Wikimedia projects (SDC support)

Data re-use between Wikimedia Commons, Wikidata and Wikisource, for digitized and transcribed books, manuscripts, archival documents...
Updated templates on Wikimedia Commons
...

Data modelling with structured data on Wikimedia Commons (SDC support)

Data modelling for GLAM-Wiki related multimedia files
...

Structured Data on Wikimedia Commons, integrated and re-used in external tools (SDC support)

Structured Data on Commons, structured copyright metadata, and the integration of Wikimedia Commons in Creative Commons Search (GitHub issue)
Hacking and development around the Attribution Generator:
- Updating the Attribution Generator to work with structured data
- Create a gadget to include attribution texts into Wikimedia Commons itself
...

Roundtripping metadata between Wikimedia sites and source databases (FindingGLAMs support) (SDC support)

Further experiments as a followup to the project Wikimedia Commons Data Roundtripping

Development with IIIF on Wikimedia projects (SDC support)

The International Image Interoperability Framework (IIIF, pronounced as 'triple-eye-ef') is a standard for sharing images (and other media files) across the web, which makes image and media repositories interoperable with each other. IIIF is widely supported by cultural institutions around the world.

Wikicite stuff!

Let's talk about journal metadata & Wikicite initiatives. Thursday meetup details to come. -- Phoebe (talk) 10:29, 14 August 2019 (UTC)[reply]

To get an idea – with examples – of what types of applications become possible with IIIF, see this blog post (June 2018) from the Wellcome Collection.

As of 2019, default IIIF integration is not on the official development roadmap for Wikipedia, Wikimedia Commons or other Wikimedia projects (such as Wikidata). The IIIF community and the Wikimedia community, however, are in touch with each other, exploring options together. Several Wikimedia community members work on IIIF prototypes that work with Wikimedia content. The Wikimania 2019 Hackathon offers opportunities to work on IIIF with the new functionalities of Structured Data on Wikimedia Commons as well. See this general info page for existing projects and ideas: Commons:International Image Interoperability Framework

Contact

Structured Data on Commons: Sandra Fauconnier sfauconnierwikimedia.org (Program Officer, GLAM and Structured Data at the Wikimedia Foundation)
FindingGLAMs: André Costa andre.costawikimedia.se (COO at Wikimedia Sverige)