Refine
Year of publication
- 2018 (2) (remove)
Document Type
- Conference Proceeding (2) (remove)
Keywords
- CVMA (1)
- Cultural Heritage (1)
- Digitale Editionen (1)
- TEI (1)
- XMP (1)
- digital editions (1)
- digital images (1)
- metadata (1)
quoteSalute strives to make data of digital scholarly editions of letters (DSELs) accessible in a playful fashion by enabling users to integrate salutations from DSELs in their own email correspondence. The foundation of quoteSalute is a curated TEI-XML text corpus which has been created by extracting <salute>-tags from TEI-XML-encoded DSELs. For providing users with fitting salutations, we annotated the data regarding language, level of politeness and intended gender of sender and receiver.
When managing large quantities of data, it is a common solution to utilize a centralized
data management software to forge a connection between metadata and the data objects
themselves. In case of text-based objects without any attached metadata, it is easy for
humans to contextualize these objects by recognizing patterns such as filenames, titles,
authors etc. This task becomes a challenge when dealing with non-text-based objects like
images in the cultural heritage domain. Without metadata or expert knowledge, it becomes
difficult to estimate the creation date of a painting or tell the name of its painter. Thus, the
ability to contextualize data depends on whether there is a working connection between
the metadata store and the data object itself. This connection fails as soon as the file is
moved on the file system without having these changes also applied in the corresponding
data base, or when the file is shared without a reference to its original location. This paper
presents an approach to overcome that type of co-dependency by utilizing XMP to embed
cultural heritage metadata directly into image files to ensure their location-independent
long-term preservation. The “Corpus Vitrearum Medii Aevi” Germany (CVMA) project
serves as an example use-case.