The Time Machine FET Flagship builds a Large Scale Simulator mapping 2000 years of European History, transforming kilometres of archives and large collections from museums into a digital information system. These Big Data of the Past are common resources for the future that will have a huge cultural, economical and societal impact. Researchers from all over the world are now joining forces to bring the past back in one of the most ambitious project ever on European culture and identity.
Cultural Heritage is one of Europe’s most precious political, economic and social assets. Today, Science and Technology can profoundly transform the conservation and experience of the cultural heritage with an impact on research, education, new applications and, in turn, on the economy and society at large. Computer and Data Sciences, Physics and Chemistry, Material Sciences and Robotics, can join forces with the Humanities to open a new paradigm for Historical Sciences.
The Time Machine FET Flagship aims at building a Large Scale Historical Simulator mapping 2000 years of European History. Extending on the proposal submitted to the attention of the European Commission in April 2016, Time Machine is a program that brings together research teams from all over Europe and the participation of about 200 institutions. The goal of this consortium is to develop new technologies for the scanning, analyzing, accessing, preserving and communicating of cultural heritage at a massive scale. Data extracted from this digital patrimony are the basis for the reconstruction of the historical evolution of most European cities and the economical, cultural and migration networks between these urban nodes.
This is something of complexity and scale unseen to date. To obtain the necessary data for such a reconstruction, Time Machine has to develop new technologies for a scanning infrastructure able to digitize massive amounts of fragile documents from the European heritage that would be the basis of the largest database ever created for European archival documents. Meanwhile, high performance computing clusters are used to process this mass of documents using increasingly accurate machine vision algorithms, segmenting, indexing and transcribing their content, ultimately making them searchable like any other documents we search on the web. The information networks extracted from the documents constitute a massive semantic graph of linked data – probably the largest ever built about the past - unfolding in space and time as part of an historical geographical information system.
These Big Data of the Past are expected to lead to data-driven historical simulations, making the past de facto as easily accessible as the present. New families of historical search engines, as well as immersive and augmented reality interfaces and other tools, will generate what one could describe as time capsules to seamlessly navigate 2000 years of European history. Thousands of time travellers are already ready to engage in the project, for curating the data, design algorithms and ultimately, to a certain extent, writing a common history of Europe. That approach in turn could apply to other cities, communities and regions of the world.
Time Machine is anchored in the technologies and methodologies pioneered by the Venice Time Machine, an on-going 10-year seed project focusing on the city of Venice and its 1000 years of history and featuring EPFL, Università Ca’ Foscari, Archivio di Stato di Venezia, Fondazione Giorgio Cini, as well as an international board including scholars from Princeton, Stanford, Columbia and London Universities. Venice Time Machine provides a proof of concept of archival digitization and machine learning to reconstruct the shape of the city over its history, along with the evolution of its continental and Mediterranean Venetian networks over time. The project maps circulation of news, money, commercial goods, as well as migration of artistic patterns along the roads from Venice to the Netherlands and Germany or down to ports of the Black sea, reconstructing, through the history of Venice, a united story of the construction of Europe. Time Machine is a spatial and temporal extension of this ambition, the potential of which the Venice chapter already highlights.
All technological development of the Time Machine FET Flagship are open source, in line with the notion of a shared patrimony and cost effectiveness for other institutions in following the same methodologies and encouraging the creation of start-up and services based on similar approaches and standards. That can lead to local initiatives for fostering tourism, cultural entertainment services and new approaches to urban planning.
We believe Time Machine and the projects it will generate represent a unique opportunity to take Europe’s cultural heritage to the future, enhanced as a shared patrimony and a common history, for the next generations.
The Time Machine FET Flagship is committed to valorize the Europeana integrated catalog and collections, as the basis for higher level data models. At the technical level, Time Machine will help the national aggregators and partners to leapfrog to new technological standard for cultural heritage storing, display and analysis, facilitating the access and enrichment of Europeana’s 53 million cultural documents.
The Time Machine FET Flagship has the support of the Archives Portal Europe Foundation, a consortium of 17 European national archival institutions, which maintains and further develops the Archives Portal Europe, the largest online access point to descriptions of archival material. At the moment – February 2017 – the Archives Portal Europe gives access to more than 250 million descriptions brought together by more than 800 archival institutions from 33 European countries. To these descriptions are already more than 200 million digital objects (scans) directly linked via the Archives Portal Europe and much more are made accessible, namely via their presentations on the websites of the Archives Portal Europe’s data providers. The Archives Portal Europe team has recently started an ambitious endeavour: providing a uniform way of publishing more detailed information from rich archival sources such as Notary records. For these so-called additional finding aids the newest version of the international standard for archival descriptions EAD will be used: EAD3, which can tag detailed information on for instance events, persons and places in a more meaningful and semantic way. This is exactly the kind of information the Time Machine FET Flagship project needs and the Archives Portal Europe will offer the project access to its network of data providers and co-operation in implementing ways to make use of their information via its infrastructure.
The Time Machine FET Flagship receives the support of DARIAH, the pan-european research infrastructure for the Arts and the Humanities, regrouping several hundreds of scholars and dozens of research facilities. DARIAH’s working groups covering topics ranging from Text and Data analytics to Guidelines and Standards are already playing a crucial role for organising the technological and scientific discussion addressed in the Flagship and will be key to structure their on-going development at the European level.
The Time Machine FET Flagship is strongly supported by the EuroSDR – the European Spatial Data Research network linking national mapping and cadastral agencies with research institutes and universities for the purpose of applied research in spatial data provision, management and delivery. EuroSDR advanced expertise about 3D modelling, spatial data archiving, historic data infrastructures, advanced photogrammetry, Linked Data, Big Geodata and ‘Geo’-business modelling contributes to making the Time Machine reality.
The Time Machine FET Flagship is supported by the ICARUS network, regrouping 160 archives and scientific institutions from 30 European countries. This continuously expanding network of coordinated institutions brings to the project the access to extremely large collection of documents and the access to relevant expertise to interpret them. Time Machine shares equally with all these institutions the technological innovations and infrastructure developed in the Flagship.
The Consortium of European Research Libraries is partner of the Time Machine Project. It includes 300 European libraries (and a growing number of US libraries with substantial European book heritage) and run the Heritage of the Printed Book database, with over 6 million records of books printed between 1450 and 1830.
Time Machine promotes the use of the IIIF APIs, guidelines and best practices. It encourages partners to share their cultural heritage data using these open standards and to develop new services in this framework. Time Machine also organises the development of higher level services based on the IIIF standards including handwritten word recognition, image matching, document analysis, complex annotations and more and explores the extension of the IIIF approach to other kind of services, beyond image delivery. This could include services for music, video, 3d, maps, biography, genealogy, and so forth.
The Time Machine FET Flagship is supported by the READ H2020 project, a multidisciplinary consortium of 13 partners working in Computer Science, Pattern Recognition, Machine Learning, Image Processing and Humanities. The project boosts the development and usage of cutting edge technology for the automated recognition, transcription, indexing and enrichment of handwritten archival documents and develops a collaborative platform for this purpose.
Time Machine collaborates with the European Association for Urban History (EAUH). EAUH was established in 1989 with support from the European Union. Its goal is to provide a multidisciplinary forum for historians, geographers, sociologists, anthropologists, art and architectural historians, planners and other scholars working on various aspects of urban history from the middle ages until the present. The EAUH supports the FET Time Machine Flagship because it will be an excellent opportunity to make big data available and accessible for the reconstruction of the historical development of European cities in relation to economic growth, cultural diversity and migration networks. These topics are highly relevant to the scientific network and the conferences of the EAUH.
Time Machine collaborates with the European Research Infrastructure for Heritage Science (E-RIHS). This pan-European project aims to support research on heritage interpretation, preservation, documentation and management. The project will provide state-of-the-art tools and services to interdisciplinary research communities that advance understanding and preservation of global heritage.
Time Machine collaborates with the MAPIRE project, a cooperation between the Austrian State Archives, the Geophysics and Space Science Department of the University of Budapest and a number of other institutions. The project originally enabled the user to navigate historical maps of the Habsburg empire using state-of-the art technologies including Google Maps, Google Earth and OpenStreetMap. It recently extended its area of coverage to now include 20 European and more than 100.000 map sheets.
Time Machine collaborates with 15cBOOKTRADE to help retracing the history of 450 000 copies of incunabula, currently located in about 4000 different public libraries. The circulation of these books will be informed by the other historical reconstructions and circulation maps developped by the Time Machine’s partners.
Time Machine collaborates with Fragmentarium, the International Digital Research Lab for Medieval Manuscript Fragments. Collaborating with 15 partner institutions throughout Europe and the USA, the project aims, over the next years, to lay the foundations for research on medieval manuscript fragments by providing open standards and guidelines.
Time Machine works with CLARIAH, the Common Lab Research Infrastructure for the Arts and Humanities, to define sustainable strategies for accessing, processing, preserving and exploiting digital data for humanities scholars.
Time Machine collaborates closely with the Golden Agent project, which goal is to disclose existing and new datasets on the production of the creative industries of the Dutch Golden Age by combining linked open data and multi-agent technology. The project will make 2 millions scans available of notarial acts/probate inventories in the City Archives of Amsterdam providing this way insight in the consumption of cultural goods of the 17th Century of Amsterdam.
Time Machine collaborates also with the CREATE Program of the University of Amsterdam. This program volunteered to develop an Amsterdam Time Machine, based on the Venetian experience, exploiting the wealth of data and knowledge present within the Amsterdam context.
Time Machine collaborates with POSTDATA (Poetry Standardization and Linked Open Data), an ERC Starting Grant EU Funded project, whose aim is to transform poetry collections into standardized, machine-readable Linked Open Data. It will allow researchers from different languages and cultures to retrieve information from sources of varied origins, making data interoperable thanks to the combination of ontologies, Natural Language Processing technologies and machine learning applied to poetry analysis.
Time Machine collaborated with project FRIDA, a pioneer digital Atlas of festivals in Renaissance Italy, with the long-term goal to focus on festivals in Renaissance Europe, developed at the University of Minnesota, in collaboration with the researchers at Politecnico di Milano and Columbia University. FRIDA aims to develop new methodological and experimental approaches to overcome recovery biases of historical records and systematically recompose the montage of ephemeral events in a way that allows for the reconstruction of a full multi-media experience. The technological development proposed by Time Machine will substantially advance the tool set available to interrogate the FRIDA Atlas.
The Time Machine FET Flagship receives the support of EUscreen. The EUscreen portal was built by a consortium of European audiovisual archives, public broadcasters, academic and technical partners. The EUscreen project, funded by the European Commission, ran from 2006-2009 under the name Video Active, and from 2009 - 2012 as EUscreen under the FP7 programme. Its successor, EUscreenXL, began in 2013 and ran until 2016 under the CIP ICT-PSP support programme. The portal offers free online access to thousands of items of audiovisual heritage and is the main hub for European audiovisual content. The project continues as the EUscreen Foundation.