Documentation
Migration project
The project has a long history and can be classified as a pioneer of electronic publication (see also the information on the project history). The necessary infrastructure was built up and developed over the years at the Institute of Linguistics at the University of Graz by different technicians. In the course of time, it became clear that the permanent maintenance of this infrastructure was no longer manageable due to the software landscape that had grown over decades. Therefore, with financial support from the Faculty of Humanities, a project was launched in 2018-2019 that aimed to migrate the dataset into the repository Humanities Asset Management System (GAMS) of the Institute Centre for Information Modeling. This migration necessarily involved a complete reorganization of the dataset, while preserving certain structures (URIs, permalinks). In June 2022, the migration was successfully completed with the relaunch of the new web presentation.
Technical Infrastructure
GAMS is an OAIS-compliant asset management system for the management, publication, and long-term archiving of digital resources from all humanities subjects. GAMS is based on the open source project Fedora (Flexible Extensible Digital Object Repository Architecture) and is continuously developed within collaborative projects to take specific research interests into account. Since 2019, the repository has been certified with the CoreTrustSeal as a trusted digital repository. GAMS is also registered with the Registry of Research Data Repositories. More information about the repository can be found here.
Persistent identification
In the repository, all digital objects are stored under a persistent identifier (PID)
in the form of a project-specific permalink (for this project o:hsa.*). In addition, the
GAMS infrastructure includes a handle server so
that the digital objects still have an externally registered identifier (for this
project 11471/518.10.*).
Due to the migration context of the project, it was
necessary to systematically transfer already existing URIs from the old infrastructure
to the new one. All objects are also accessible under their previously used
identifiers.
Examples:
Letter: http://schuchardt.uni-graz.at/id/letter/1133 -->
https://gams.uni-graz.at/o:hsa.letter.1133
Person:
http://schuchardt.uni-graz.at/id/person/1012 -->
https://gams.uni-graz.at/o:hsa.persons#P.1012
Bibliographical work:
http://schuchardt.uni-graz.at/id/publication/181 -->
https://gams.uni-graz.at/o:hsa.bibliography#BIBL.181
Standards and data modeling
The letters are modeled in XML format, taking into account the guidelines of the Text Encoding Initiative
(TEI). The implementation is based on the DTA-basic format, ediarum and the CorrespDesc-SIG (see ODD or Relax-NG). A special feature arises with the element hi: Since the type of
highlighting was often not marked in the data acquisition, @rend="unknown" is usually
used, and the output is in italics. The usage therefore only indicates that the marked
text differs from the surrounding text, but not in which form.
Indexes of persons
and places as well as the bibliography were also modeled as TEI, the implementation of
the indexes here is also based on ediarum or the Humboldt-Edition. viaf.org
and GeoNames were used as standard
reference files.
The keyword and language thesaurus were modeled as SKOS. Languages were additionally
mapped to lexvo.
Additionally, when uploading to the repository, RDF data is automatically extracted from the TEI based on rules (see RDF-S) and written to the Triplestore. This
database is queried via the general search and enables the interlinking of
resources.
Reuse of the data
With regard to the stored data, we follow the FAIR data
principles: „data should be Findable, Accessible, Interoperable, and
Re-usable“.
All XML data and digital copies from Graz University Library are
licensed under the Creative Commons BY-NC license. Deviations only occur in
the case of digitized material from other institutions, in which case the deviating
usage information in the image display must be taken into account.
All image files
in the repository can also be delivered via the IIIF-compatible Image Server. The following syntax must be used for this
purpose:
https://gams.uni-graz.at/iiif/{object PID}/{image PID}/{IIIF query}
Example: https://gams.uni-graz.at/iiif/o%3Ahsa.letter.1779%2FIMG.1/300,500,600,200/full/0/default.png.
To systematically harvest data or integrate it into your own
contexts, the OAI-PMH interface
can be used at https://gams.uni-graz.at/oaiprovider and queried according to the PMH rules.
Correspondence data is additionally sent to the aggregation service CorrespSearch (see
the CMIF-File).
Using the web interface
The core of the project is, besides the provision of primary and secondary literature
as well as contextual information on Schuchardt, the processing of his extensive
correspondence. Much of this is already available as edited full text with facsimiles.
The letters are enriched with information on persons, places, works, keywords and
languages. The letters can be accessed via the full view of the respective
correspondence, the structured letter search or the general search; in addition, the
indexes under the item List of topics offer a further entry point to the data.
The source data in TEI or SKOS format are linked in the presentation in each case and can be viewed or downloaded there.
Contact
For questions regarding the technical implementation of the project, please contact Elisabeth Steiner (elisabeth.steiner@uni-graz.at).
Text: Elisabeth Steiner & Sanja Sarić