1 of 5

Two Centuries of Indian Print

Digitised rare and unique printed books from the Library’s South Asian collections, 1713 to 1914 to enhance study and understanding of South Asian book history

Over 2,000 items in Bengali, Sylheti, Assamese and Urdu, now available online

Work strands: digitisation, cataloguing, community and public outreach and skills exchange, research and digital research (DH)

Access books online: bl.uk/early-indian-printed-books/collection-items

bl.uk

Two Centuries of Indian Print is a major project by the British Library to digitise rare and unique printed books from its South Asian collections, dating from 1713 to 1914.
Launched in 2015, the project was funded by the AHRC, Newton-Bhabha Fund and the Department for Business, Energy and Industrial Strategy.
Digitised 1,000 rare and unique early printed Bengali books + 600 books printed in Assamese, Sylheti, Urdu and Bengali
The project covered several work strands, including digitisation, cataloguing, community and public outreach, research engagement and digital research activities

Partnerships & Collaborations

The project has been working in partnership with Jadavpur University, who have provided subject matter expertise, hosted digital workshops and contributed to the digital research work, particularly the transcription of books used to train text recognition systems.
The project also worked in partnership with the Srishti Institute of Art, Design and Technology, especially around the delivery of digital skills workshops and for the content of the project website.
The Library signed an agreement with the National Library of India to undertake a programme of collaborative training workshops and work placements, to foster skill sharing and professional links between library professionals in the UK and India.
We also worked with the School of Oriental and African Studies (SOAS): 49 rare Bengali books from their collection have been digitised and made available through the SOAS Digital Library.

Digitisation & Cataloguing: In addition to the scanning and cataloguing of 1,600 titles, we also scanned over 10,000 early printed books, mainly in Bengali and other South Asian languages; as well as 250 volumes of Quarterly Lists, which were series of publications recording all books published in India on a quarterly basis.

Two Skills Exchange Programmes were organised in 2017 with the National Library of India to engage and promote capacity-building and skill-sharing elements between the two institutions.

The project has had a rich programme of seminars, community events, show & tells, and publications;
Including a two-day symposium at Jadavpur University and two workshops taking place at the Library.
These helped create an international research network of contributors to the project’s research outcomes.

Public and Community Outreach:

The project hosted regular visits from grass-roots organisations from diaspora communities, local groups and official visits, to discuss the project and our South Asian collections, and to introduce some collection items.
To date, the project has organised dozens of public seminars on subjects related the South Asia collections at the Library, engaging researchers and members of the public. Recordings of these seminars are available as podcasts.

2 of 5

Why did the project use Wikisource?

Established connection with Bengali Wiki community through workshop

Increase visibility and widen access to 2CIP books, especially in South Asia

Wanted to make use of existing OCR transcriptions

Strengthen relationship with Indian Wiki community

Learn new Wiki skills

bl.uk

Connected with this community through an unrelated workshop we ran in India where we met the Centre for Internet and Society, who introduced us to people working on Bengali Wikisource

Also attended talk by NLS about the chapbooks they have put on Wikisource. They spoke about how they were receiving more visits than the same items on their own library website. This piqued our interest and as we had public domain books we knew we could put them up.

We wanted to increase visibility of our books. Many potential users of our books may not use the BL website as a first point of call, and especially in South Asia they may use Bengali Wikisource, especially for the Bengali interface, so we decided to put them on Wikisource as well as on the BL website to maximise the online presence of these books.

We had existing OCR transcriptions (from Transkribus) and wanted to try importing those first into Wikisource. However, it was the preference of the team working on Bengali Wikisource that we use the Google Cloud Vision API built into Wikisource, to OCR the books.

We wanted to create a relationship with the Indian wiki community and we wanted to learn new skills in using the Wiki platforms.

3 of 5

Wikisource Proofreading Competitions 2021

Diverse range of notable works selected

BL uploaded books to Wikimedia Commons > Wikisource

Promotion through blog post, social media and creation of competition page on Wikisource

Competitions for 2CIP Books Mar-Apr and Sep-Nov 2021

- fully proofread >40 books >5,000 pages

- competition prizes for top 5 contributors

Contributed books towards the Tagore proofreading competition

bl.uk

We decided to select a cross section of books from those we have digitised. Bodhi selected the books to prioritise for upload.

Once books were shortlisted we (BL) used Pattypan to batch upload multi-page PDFs to Wikimedia Commons. Bodhi created templates in Wikimedia Commons that we used during upload for certain metadata fields, specifically ‘source’ and ‘licence’ fields. This created a smooth upload. Bodhi then indexed the books to create a Wikisoure entry and he copied them over to the competition pages

Results – Through the competitions, more than 40 books/5,000 pages were fully proofread by volunteers. The competitions were limited to India due to restriction on prizes not being able to be sent out internationally by the Wikisource team Competition prizes included books, t-shirts, momentos and a certificate issued by the BL.

We also contributed several digitised works of Tagore from the BL collection to the Tagore proofreading competition that was organised on Bengali Wikisource in late 2021.

4 of 5

Upcoming Plans with Wikisource

Ongoing upload of books to Wikimedia Commons, Wikisource and Wikidata
Sharing experiences of other text recognition tools and services
Connecting WMF with cultural heritage institutions in South Asia for training events

bl.uk

GLAM page records notable activity and script for accessing images from BL

Books are still being uploaded from our collection to Wikisource by Bodhi and the BL, but due to lack of resource this is quite slow and we do not have any new ventures planned with the Indian Wiki platforms in the short term, but would like to eventually create more Wikidata entries for the books

We have also provided input to discussions with Wikimedia Foundation sharing our experience using Transkribus and how that could be integrated into the Wikisource platform to enable users to process OCR within Wikisource using Transkribus. It could potentially open up text recognition for handwritten documents on Wikisource.

We are also helping WMF to connect with archives, libraries and other cultural heritage institutions in India who we have previously run training events with, to see if they would be good candidates for Wiki related training.

1 of 5

2 of 5

3 of 5

4 of 5

5 of 5