READ project’s final meeting and the Transkribus seminar in Helsinki

A final group photo of the international READ project staff at the National Archives of Finland on June 27^th.

The last couple of weeks have kept us especially busy at the National Archives. As you may already know, READ as an EU funded project has now reached the finish line, and this has meant both summarizing the achievements so far as well as turning the focus on the future. Right after midsummer, starting from Wed the 26^th of June, NAF gathered our international READ partners to Helsinki for a three-day meeting and seminar spree. This provided the opportunity to thoroughly discuss the project’s three-year journey among the partners, but also to present some of its most central results to the public.

The meeting days kicked off with a Transkribus Seminar that was open for the public to attend. There were around 100 attendees total, and the seminar guests represented a diverse set of organizations ranging from libraries, museums and archives to universities, information technology companies, literature societies, etc. It was great to see such an active attendance and the growing interest towards computer-assisted enrichment of cultural heritage.

NAF's READ project group.

The seminar day consisted of presentations on the technologies developed in the READ project, such as Transkribus and HTR, Keyword Spotting, segmentation tools and ScanTent, as well as different archival projects that have been utilizing these tools in practise. NAF’s READ project team presented the progress we have made in transcribing 19th century court records and putting them online in searchable form. These topics are discussed in further detail in our other blog posts. (e.g. Court Records and New Search Engines).

Among the presentations were also some other examples of READ technologies in similar use as at the National Archives of Finland. The Amsterdam City Archives have used Transkribus to transcribe the Amsterdam notarial archives. The correction of the transcriptions has been crowdsourced using the VeleHanden (“many hands”) platform. The archive contains lots of information about Amsterdammers and their international networks from the years 1578-1915. The documents tell about trade, shipping, slavery, personal property and heritage, including eyewitness reports of hardships at sea, as well as ordinary neighbourly quarrels or fights in hostels and illegal playhouses. The Swedish Literature Society in Finland (SLS) in turn has used Transkribus to transcribe the famous Finnish painter Albert Edelfelt’s letters (in Swedish). The letters offer a window to the golden age of Finnish art, and tell about Edelfelt’s work, travels and meetings with other people. Like us, SLS also aims to integrate Transkribus on a larger scale in the working process of their archives and publishing unit.

Mark Ponte and Jirsi Reinders discussing VeleHanden.

Maria Vainio-Kurtakko presenting the Edelfelt project.

A panel discussion concluded the seminar.

After Wednesday’s seminar, Thursday and Friday were reserved for private meetings among the READ partners. There were many interesting presentations and discussions that gave a deeper insight into the READ project as a whole – who has done what, and what are the greatest achievements, issues that haven’t gone as planned, the affected stakeholders etc. One of the most thought-provoking presentations was about a user satisfaction survey that professor Melissa Terras from University College London had conducted among the Transkribus users. The satisfaction results were promising, but there’s also a broader context about how AI is changing the archival profession, and actually the READ partners are one of the very few who can already measure both the monetary and social value of these changes. 

As the READ project has now lasted over three years, there are several important general findings and developments resulting from it. For example, there has been a huge progression regarding the HTR models and the automatic analysis of document layouts. Because of this there is less manual work in many parts of the HTR process compared to earlier times. Throughout the project many human work years have been spent in producing the training data for the HTR models. And as the project coordinator Günter Mühlberger said, this training data is the golden nugget of the project and “sharing is caring” – there should be strong encouragement to share the training data with other participants. 

In addition, an important notion is that even though READ now ends, the work still continues in a slightly different form. On 1st of July 2019 the READ project turned into a European Cooperative Society (SCE). READ-COOP serves as the basis for sustaining and further developing the Transkribus platform and other READ-related services and tools. The National Archives of Finland has already joined the COOP and will collaborate with the different members throughout the Making of a Modern Archive project.

We will keep you posted as the work proceeds!

(More about the seminar at the National Archives' homepage (in Finnish))

Sampo Viiri & Laura Kukkohovi