*|MC:SUBJECT|*

As this newsletter finds you, the ELE project consortium finds itself in the midst of a series of events: With three conferences held in October and four more coming up in November, the initiative for digital language equality is presented and discussed all over Europe – and even in Brazil and Russia. In the first week of October, the 18th Conference of the European Federation of National Institutions for Language (EFNIL) saw contributions from ELE coordinator Prof. Andy Way (ADAPT Centre, Dublin City University) who presented the project framework on site, and Dr. Anželika Gaidienė (The Institute of the Lithuanian Language), who joined in via ZOOM to discuss the “Lithuanian Language Technology Landscape: from Documents to Language Technologies” based on preliminary results of the ELE research. In her presentation, Dr. Gaidenė shared some early insights such as the large amount of lexical and conceptual resources on Lithuanian, which make up 62% of all resources, and the strong language-dependency of tools, 98% of which are only available in Lithuanian.

Every two years the Directorate General for Translation of the European Parliament organises a conference. At this year’s DG TRAD Conference, which took place on 27 and 28 October, the main umbrella topic was Machine Translation. Prof. Georg Rehm (German Research Center for Artificial Intelligence, DFKI) was invited to present the projects ELE and ELG in the context of “Language technologies for a multilingual Europe 2020-2030”. Jumping into the future and from a European to a national level, Georg will also present both initiatives at the Symposium 2021: “Machine-Based Cataloguing Processes” of the German National Library (DNB) taking place virtually and “mainly in German” on 18 and 19 November. His contribution focuses on the “European Language Grid: An AI platform for flexible language technologies”, but also touches on ELE.

Maria Heuschkel of Wikimedia Deutschland on the other hand has just finished presenting the preliminary ELE project results and discussing the pains of under-resourced language communities that are active in the Wikidata community at WikidataCon, the conference for “everything about Wikidata, the free, collaboratively created database of structured data” organized by Wikimedia Deutschland and Wiki Movimiento Brazil between 29 to 31 October. In the session on ELE, questions about the challenges, needs and expectations for the future of Language Technology for under-resourced languages and the role of policymakers for the preservation of European languages through Wikimedia projects were posed to the conference participants. At the upcoming CEE Wikimedia Conference from 5 to 7 November, Maria will host another session on ELE, presenting the current state of the project such as the preliminary definition of Digital Language Equality, the primary survey results and the languages involved. The conference organized by members of Wikimedia from Austria, Poland, Greece, Russia and others countries is the yearly meeting of Wikimedians from Central and Eastern Europe centred on Wikimedia projects. The ELE session will involve group discussions with community members about challenges and problems when working with technologies in and for their languages.

On to further questions: Do you know whether and how your library uses digital search tools, translation technology, speech recognition or spell checkers? And how would language equality affect such services? These and many more questions will be posed at the workshop on “Achieving Digital Language Equality 2030: Implications for Libraries, Collections, and Library Users” hosted by the Association of European Research Libraries (LIBER), partner of the ELE project. The online workshop takes place on 18 November and presents speakers from ELE and practitioners working with language technology, painting a picture ofwhat a future entailing language equality throughout Europe looks like. Participants can learn about the work of ELE and the potential of digital language equality in the library sector, but also contribute input and discuss how digital language quality would affect them. Registration is now open!

The Catalan language is known for its cultural and political importance, but only two percent of the European Union’s population is able to speak it, while less than one percent is native to the Latin language used along the Spanish and French Mediterranean coast. The gap to the Spanish national language, spoken by 17 percent of the EU’s population, is immense and prone to grow larger in the digital world, where resources are typically focused on larger audiences. To increase the technological support for Catalan, members of the ELE project working at the Barcelona Supercomputing Center (BSC) and other NLP researchers have founded a new community: NLP CommuniCat, aiming to unite and create exchange between the people working with language technology in Catalan. More than 40 participants soon gathered in the Slack chat group, while the NLP CommuniCat Twitter account gained more than 220 followers within a single month. The initiative to combine research efforts and find synergies between experts working with Natural Language Processing and Language Technology is a great example of how inequality between digital languages can be fought through a joint effort.

Selected new ELG members