Contact e-mail: digitalia@phil.muni.cz
Updated: 02/04/2026
The faculty data infrastructure Digitalia MUNI ARTS is technically based on the Islandora system, which consists of several components:
Drupal is used to manage the repository content, store metadata and provide a friendly user interface.
The files themselves are stored in Fedora. Fedora services include versioning and fixity checking.
Apache Solr is used for indexing content and searching the repository.
Islandora microservices provide synchronization of Drupal and Fedora content and integration of various image, video, and text processing applications.
A more detailed diagram of Islandora’s architecture is included in its documentation.
Apache Solr is used for indexing and searching the repository.
Characters belonging to the following categories (Unicode Character Categories) are ignored during searches:
Punctuation, Connector Characters
Punctuation, Dash Characters
Punctuation, Close Characters
Punctuation, Final quote Characters
Punctuation, Initial quote Characters
Punctuation, Other Characters
Punctuation, Open Characters
The system uses the infrastructure for running virtual servers at Masaryk University built on VMware technology located at the MU Institute of Computer Science. The operating system is Ubuntu LTS (as recommended by Islandora Community).
Backups are performed regularly to a tape device on Bacula system managed at the MU Institute of Computer Science.
Changes in metadata of all record types are recorded in Drupal. Users with the Content editor role can view, compare, delete, or revert to a previous version of objects (called Revisions).
Beside files, all record types are also stored in Fedora. This happens automatically with each update of a record. S3 compatible object store is used for persistence, items are saved in accordance with OCFL standard. This allows for recovery and migration to another system in case Islandora ceases to exist.
An AIP (archival information package) is currently the whole OCFL tree stored by Fedora, but finer control is planned in the future.
Content of databases is accessible to non-logged-in users as well. Such a user can browse individual records and files, apply filters and use search.
Logged in users with the Content editor role have access to record edit forms. They can also manage versions of objects (see Servers, backups, integrity and authenticity).
The content of the repository is accessible via the REST API. The supported formats for the GET method are csv, json and jsonld.
The Matomo tool is deployed to track the usage of the site. The collected data is stored on a local server.