Gruppo Logos

Gruppo Logos (the Logos Group) is a translation group based in Modena, Italy. They have developed a number of free multilingual projects which are among the world's largest, with the help of a base of 3000 professional translators. These projects have been underway since 1988, and have been freely available online, without advertising, since 1995. They maintain their own servers, and their own team of software developers.

Existing projects

Long-standing projects include:

• Dicologos, the Logos dictionary (7M total entries in 200 languages and dialects),
• A Children's Dictionary, with cute illustrations for each word and pronunciations for about half of them (150 languages; 800 words in english, hebrew and quechua; only 300 in rapa nui)
• A Wordtheque (a searchable corpus of almost 1B contextualized words in 100 languages [NB: adding Wikipedia to this corpus would double it]),
• A Universal Conjugator, which conjugates verbs in 35 languages; and helps you find which language you want based on a word
• A Children's Library (a collection of children's stories in 34 languages, many with accompanying audio books)
• Various toolbars integrating these databases into browsers and text editors.

Projects under development include:

• Arcoiris : A webtv archive or videotheque : a library of free documentaries and cultural films (currently 3k videos, mainly in italian but also in german, french, & english)
• Newspapers : one free newspaper (Piel de Leopardo) is under development in Spanish

Project history

See the historical press releases (in English: "dalla stampa", "communicati") for further history of these projects.

Interest in collaboration

I contacted Logos in early 2005 to ask for advice and help in finding interpreters for Wikimania; and to invite them to discuss the history of their own free web-based collaborations at the conference. They were amenable on both counts; the founder and one other plan to be at the main conference. They expressed interest in collaborating with Wikimedia to avoid duplication of effort, and to improve both collections of data. There are both data-structure and license issues to work out; their data is all structured, but without extensive metadata about users and change histories. They also strongly prefer non-commercial licenses, but are willing to discuss license options. +sj | Translate the Quarto |+

Specific possibilities for collaboration :

• Helping add Wikipedia content to the Wordtheque. Snapshots? Updates how often?
• Wiktionary coordination. They have extensive content, contributors, and subject expertise (in definitions, translations and pronunciations); but no wiki and no explicitly free license. They have already tackled the data-structure problem, with different needs. How to preserve both project interfaces while sharing content?
  • Similarly, they have programmers working on maintaining their software; could work on project and data integration, import/export issues, etc.
• TV and other content archives: how to coordinate this with wikicommons, archive.org, et. al?
• Newspapers: details about their efforts need clarification. Content sharing and cross-promotion with wikinews would be interesting.
• Hosting: Logos has their own data center, support staff, and ~40Mbps of bandwidth. If farming out squid clusters ever becomes all ther rage...