Our guiding principles are modelled on and adapted from the e-codices Switzerland website:
- Manuscripts: Our focus is western manuscripts from the Middle Ages and the early modern period (ca 500-ca 1600).
- Research library: The publication of digital manuscripts follows the criteria of scientific editing and is mainly aimed at a scientific research audience. The project aims to include other audiences in the future.
- Netherlands: eCodicesNL publishes digital manuscripts from Dutch collections on the Internet. This prototype involves three partners. The project aims to integrate institutions and collections from all parts of the country.
- Language: All static pages and basic metadata for all manuscripts are provided in English. The project aims to offer these in Dutch in the future.
- Quality reproductions: We aim for very high photographic quality according to the principles of scientific research photography.
- Quality-rich metadata and search functions: Each manuscript is linked to basic metadata, which are based on the collection of this data in MMDC: Medieval Manuscripts in Dutch Collections. We plan to improve the quality and depth of these descriptions in the future.
- Interoperability: We aim to use shared standards (IIIF, XML-TEI) to create as comprehensive a network as possible and enable automated sharing of images and metadata through different interfaces.
- Open Access: All digital reproductions and metadata are freely accessible for non-commercial use (CC-BY-4.0, https://creativecommons.org/licences/by/4.0/).
Guidelines for reproduction
- Camera, lens
The digitisation was carried out with a medium format film camera Fuji GFX 50R. Two lenses were used: Fujifilm GF 63mm f/2.8 R WR and Fujifilm GF 45mm f/2.8 R WR.
- TIFF Master Image Files
Proprietary RAW files created by the digital cameras were processed with the software package Capture One (adjusted, cropped and colour-checked) and converted into TIFF files. These master files are archived at the Huygens Institute and shared with the holding institutions. The uncompressed TIFF files are created with a resolution of at least 300 ppi and a colour depth of 16 bits and contain all the data needed for colour management.
For the viewer of the web application and for the image server, TIFF files are converted into JPEG files in order to deliver the images online with speed. The JPEG derivatives have a quality setting of 97%.
Manuscripts are fragile and unique. To capture them on camera, they were treated with care. Firstly, we chose the Traveller’s Conservation Copy Stand (TCCS 4232), a mobile camera setup specially designed for material from special collections. This allowed us to photograph on site; the books did not have to be transported. Moreover, we ensured a handling protocol that respected the age and fragility of the objects:
- manuscript specialists or internally trained assistants handled the manuscripts;
- reading cushions were used to temporarily store the books during photography;
- minimal manipulation of the manuscript itself, except for turning the pages;
- minimal contact of human skin with the written space in the book;
- the angle of the opened book did not exceed 110°;
- no mechanical pressure was used to flatten the pages, as with a glass plate;
- only the acrylic finger was used to hold the book open;
- acid-free paper was used to take pictures of pages with holes or other damage;
- no flash was used;
- LED lighting of the TCCS was used to prevent the temperature from rising to undesirable levels and changing humidity.
Coding of metadata
Manuscript descriptions were encoded in a specialised XML format, which complies with the Text Encoding Initiative guidelines for the electronic presentation of manuscript descriptions (TEI-P5).
Compliance with this international standard enables the exchange of metadata with other digitisation projects, meta-search engines and portals. Moreover, structuring the metadata according to the TEI-P5 standard makes it possible to assign metadata information to individual fields or substructures, allowing users to perform targeted searches in the database, e.g. by author, title of the work, date or place of origin, decorative elements, and so on (more information in “how does it work“).
Moreover, the use of XML guarantees the long-term usability of the current metadata, independent of any particular software or edition of that software.
Every digitised manuscript is accompanied by basic metadata. These minimal metadata were collected from the Medieval Manuscripts in Dutch Collections (MMDC) project, which coded them in a Dublin Core standard. The codes were translated into TEI labels from the TEI-P5 scheme and rewritten. Obvious gaps or errors in the data were corrected, but generally the integrity of the MMDC metadata was kept intact. All descriptions on our website follow as much as possible (as far as reasonable and prudent) the descriptions in MMDC on which they are based. A provision to allow users to access and download the XML format/TEI-P5 versions of all online descriptions is under construction.