The goal of LT-Web is to set the foundation for the integration of language technologies into core Web technologies, via the creation of a standard defining three kinds of metadata about: 1) information in Web content being relevant for language technology processing; 2) processes for creating Web content via localisation and content management work flows; 3) language technology applications and resources used in these applications. The organizations participating in LT-Web will join the World Wide Web Consortium (W3C) and create a group defining the metadata standard, within the Internationalization Activity of W3C.
Using the credibility and outstanding position of W3C, LT-Web will create broad consensus across communities, involving producers of content, localization workers, language technology experts, browser vendors and users. W3C will help to establish a network between stakeholders in these communities, to ensure the long-term adoption of the metadata standard and future opportunities for integrating language technologies and emerging Web technologies. Finally, LT-Web will lay the technical foundations for new business opportunities, to reach out to a growing linguistic and cultural diversity of Web users in the European Union and its trade partners, and to respond to their specific needs in a timely and cost-effective manner.

This project will deliver a modular software platform that will enable SMEs in the localisation industry to leverage open language resources published by public bodies. This platform will bootstrap a commercially sustainable localisation process that allows SME Language Service Providers (LSPs) to continuously reuse the output of web content translation projects. By integrating these outputs with language resources from public bodies, SME LSPs can easily and cheaply optimise and tailor open source machine translation and automated terminology extraction components to the needs of their customers. Today, the curation of parallel text involves the publication of the undifferentiated results of translation projects in differing formats or via centralised repositories. This restricts SMEs in leveraging these resources to train machine translation engines to match incoming translation projects.

FALCON will enable the controlled sharing and reuse of language resources, combining open corpora from public bodies with richly annotated output from commercial translation projects. Federated access control will enable sharing and reuse of commercial resources while respecting business partnerships, client relationships, competitive and licensing concerns. Fine grained traceability language resource reuse will enable return on investment calculations on resource curation in public bodies and SME LSPs alike. The consortium combines academic expertise in localisation interoperability, linked data management, statistical machine translation and statistical parsing with SMEs offering services and solutions for translation and terminology management. This project will enable the language service industry to realise its potential as a major sources of shared linked data by measurably improving cost, speed and quality benefits of reuse. The consortium members will then be strongly placed to commercially exploit these benefits.