Projects: “Time Machine: A Pan-European Digitisation and Processing Infrastructure”
From the Time Machine Project:
…the EU-funded Time Machine CSA: a pan-European initiative for digitising and processing cultural heritage assets across Europe. The project currently involves around 300 European institutions from 32 countries including: 7 national libraries, 19 state archives, some of the most important museums in Europe (the Louvre and Rijksmuseum), as well as hundreds of other research institutes, companies and governmental bodies.
Currently, Time Machine is working on two fronts: the definition and development of its distributed infrastructure, and the standardisation of operations between the project stakeholders. Joining Time Machine’s network ensures that museums, archives, libraries, monuments and other cultural heritage sites can benefit from the digitisation capabilities of all Time Machine operators. These competencies range from robotic-based mass digitisation pipelines for documents and objects, to drone-based scanning systems. Time Machine’s infrastructure represents a pan-European platform that will be used not only for matching digitisation needs with digitisation services, but also for channelling funding opportunities to cover digitisation costs and for developing services through platforms dedicated to a variety of domains (e.g. tourism, creative industries, smart cities, land-use policies, etc.).
For partners in Time Machine’s network, digitisation is only the first step in a long series of extraction processes. These include document segmentation and understanding, alignment of named entities and simulation of hypothetical spatiotemporal 4D reconstructions. Time Machine’s infrastructure is currently working on standardised and generic deep learning-based modules that can be composed in sequence to extract information based on state-of-the-art technology. Any document using Time Machine standards will benefit from a full series of extraction tools, including the latest handwriting recognition technology and document analysis algorithms. The commonality of these modules also allows us to envision a distributed processing solution hosted by Time Machine partners, which are deployed based on the territorial progress of the digitisation program.
Since its launch, Time Machine has received offers from hundreds of volunteers wishing to help digitise and valorise European cultural heritage or to participate in the technical development of this pan-European infrastructure. Time Machine coordination units are now developing solutions to help structure this community of communities. Local inhabitants of a city will have the possibility to help model their own neighbourhood or to bring their own private archive to the Time Machine. Genealogists will find potentially missing elements within the new stream of documents and data for reconstructing historical links between the ancestors they study. As Time Machine codes are open source, developers can also join in by upgrading elements of the infrastructure.
Direct to Time Machine Website
Direct to Time Machine on Twitter
About Gary Price
Gary Price (firstname.lastname@example.org) is a librarian, writer, consultant, and frequent conference speaker based in the Washington D.C. metro area. Before launching INFOdocket, Price and Shirl Kennedy were the founders and senior editors at ResourceShelf and DocuTicker for 10 years. From 2006-2009 he was Director of Online Information Services at Ask.com, and is currently a contributing editor at Search Engine Land.