QUEERLIT database: Metadata Development and Searchability for LGBTQI Literary Heritage
An investigation of LGBTQI in Swedish literary history is hampered by the fact that to date, there are no broad scholarly inventories of such literature. Research on indexation has also revealed problems in relation to representation. The systems and thesauri in use are not fitted for describing LGBTQI themes, motifs, or characters. The purpose of this project is twofold: to develop methods for the identification and indexation of LGBTQI literature, serving to enable further research within the field, and to create a subdatabase of LGBTQI literature in LIBRIS. The project has four specific aims: 1) developing a thesaurus for indexing LGBTQI literature in collaboration with KvinnSam, and mapping this to existing indexing systems 2) identifying LGBTQI fiction in collaboration with an advisory board consisting of experts in Swedish LGBTQI literature 3) constructing a subdatabase in LIBRIS containing bibliographic records of LGBTQI literature in collaboration with the Swedish National Library and KvinnSam 4) making the subdatabase available through a separate interface allowing for more specialized searches than LIBRIS does and linking the entries to open data. Since LGBTQI in historical texts is often expressed indirectly or with other concepts than the current, the project will also deal with the methodological problems arising from the application of modern subject headings to older literature.
Final report
PURPOSE AND DEVELOPMENT OF THE INFRASTRUCTURE
The project aimed to achieve three key objectives:
1. Create a bibliography for Swedish lgbtqi literature in Libris.
2. Develop a thesaurus for improved subject description of this literature.
3. Design an external interface for the bibliography that offers enhanced search capabilities.
Queerlit provides new opportunities to identify literature that either represents lgbtqi experiences or has been interpreted as queer in research or queer communities. The Queerlit bibliography includes fictional works that portray same-sex sexual practices, emotions, and/or identities, as well as depictions of transgressions of binary gender norms through body, clothing, and/or identity. The project established inclusion criteria available on Queerlit’s website, applying these systematically. The bibliography incorporates works that explicitly address LGBTQI themes as well as those where such meanings are evident in context. This “context criterion” captures the history of lgbtqi literature, where such themes were often implicitly rather than explicitly represented.
The QLIT thesaurus was developed based on the largest international controlled vocabulary for LGBTQ+ materials, the Homosaurus. Terms were translated and adapted to Swedish contexts, and mapped to Homosaurus, Swedish Subject Headings (SAO), Children's Subject Headings, and Library of Congress Subject Headings (LCSH). These terms were used to enhance the subject descriptions and searchability of works included in the Queerlit bibliography.
The Queerlit interface provides better search capabilities than Libris web search. It was launched in February 2023 and developed iteratively through workshops and feedback from the indexing team. The interface development and linked thesaurus were created in dialogue with the National Library of Sweden (KB), addressing integration of thesaurus data into Libris and adapting the Libris search API for Queerlit’s needs. The interface enables term-based searches and visualizes the hierarchical structure of the thesaurus, showing relationships between broader and narrower terms.
PROJECT OUTCOMES
The Queerlit bibliography currently contains 2,375 titles in Libris, surpassing the initial goal of 1,000 titles. It spans works from the 6th century to the present. Materials were identified through prior research, subject term searches in SAO and the Children's Subject Headings, full-text searches in the Swedish Literature Bank and Project Runeberg, student theses, tips from the public and librarians, social media, community workshops, and monitoring of new publications. Literature published before 2000 was accompanied by descriptive texts justifying its inclusion.
By the end of the project, the thesaurus contained 913 terms, with 779 matching Homosaurus, 208 matching SAO, and 61 matching Children's Subject Headings. However, 120 terms lacked equivalents in any of these systems. Each term includes a scope note defining its use. A list of 81 terms deemed broad enough for SAO was submitted to KB and integrated into SAO.
Indexing in the bibliography diverged slightly from traditional fiction indexing. QLIT terms were pre-developed, with new terms added iteratively during indexing. A “user warrant” approach prioritized terms likely sought by users, and exhaustive indexing was applied, deviating purposely from Libris norms. QLIT is an openly available resource, with its subject terms linked in Libris. Indexing guidelines were developed in collaboration with KB and KvinnSam.
User testing with librarians and other users was utilized in the interface development. Research on improving interfaces to leverage underlying knowledge organization informed the visualization of thesaurus term relationships.
RESEARCH INITIATED WITH THE INFRASTRUCTURE
The project established an advisory board with subject experts and a Libris representative. This group supported strategic decisions regarding inclusion criteria, the interface, and thesaurus development. Collaborative proposals with national and international researchers were submitted to funding bodies like VR and Chanse Hera, although not funded. These application collaborations contributed to establishing contacts with scholars focused on minority information access in European libraries and digital cultural heritage infrastructures useful for further research collaborations. Queerlit also cooperated with the Nordic Network for Queer Archives and Activities (NNAQH) and the Dawit Isaak Library on a database of banned books.
The search needs of users were examined in two studies in 2022 and 2023, one directed at general users and one at librarians. The results demonstrated the need for an infrastructure like Queerlit in both groups. Users lacked specific terms and found the search interfaces to be user-unfriendly. Professional subject index terms assigned by information specialists using the QLIT thesaurus have also been compared to the performance of automated solutions like ChatGPT and Annif, and QLIT has been compared to social tags.
CHALLENGES
Queerlit has served as a pilot project to test the new possibilities in Libris XL, which supports linked open data. QLIT is the first external subject heading system that can be used in Libris, and this work has tested both the possibilities of making external data available for import into Libris and using the Libris API to display the bibliography through an external search interface. It was not possible to fully implement all desired search functions due to limitations in Libris’s software platform. However, it should be emphasized that through the dialogue with Queerlit, the need to modify LibrisXL’s API was identified, which may also benefit other projects.
MANAGEMENT AND ACCESSIBILITY
After the project's conclusion, KvinnSam, according to an agreement between the Gothenburg University Library and the fund manager, is responsible for indexing new titles in the database based on suggestions from a working group. The working group consists of four members from the Expert Network for LGBTQ+ Issues within the Swedish Library Association, including a former member of Queerlit. The working group also meets with a representative from KvinnSam twice per year.
An editorial team for the QLIT thesaurus has been established, consisting of four members from Queerlit, one of whom is employed at KvinnSam, as well as a representative from SAO. Indexing guidelines have been developed and are openly published on the project’s website, as well as linked from Libris. GRIDH, Gothenburg Research Infrastructure for Digital Humanities, is responsible for ensuring that the external interface functions and is updated. The code for the interface is openly available on GitHub.
ITERNATIONAL COLLABORATIONS
The chair of the Homosaurus editorial board has acted as an advisor to the project and as a contact person for interoperability between QLIT and Homosaurus. Queerlit has submitted a list of false matches between LCSH and Homosaurus identified during the work with QLIT, as well as suggestions for corrections to the hierarchical term structure. Thus, Queerlit has contributed to improve the quality of Homosaurus. Work to translate Homosaurus into various languages has been initiated, and in 2024, the translation of terms in Homosaurus into Swedish began in collaboration with Queerlit. As of 2025, a project member has joined the permanent editorial board of Homosaurus.
The Scientific and Technical Advisory Council (STAC) at the International Society for Knowledge Organization (ISKO) was established as a result of the Queerlit project to create international guidelines for end-user interfaces based on subject metadata available in many library catalogs but rarely utilized at the interface level. Search functionalities proposed by ISKO have been tested and implemented to the extent possible in the Queerlit database and will be used as ‘best practice’ examples for various functionalities.
Queerlit has also given rise to research on subject metadata in search systems in collaboration with the International Society for Knowledge Organization (ISKO), and discussions have begun with the International Federation of Library Associations and Institutions (IFLA) regarding international guidelines for subject indexing of lgbtqi resources.
The DARIAH-EU infrastructure for digital humanities has a platform for open educational resources, DARIAH Teach. In the course “Introduction to Knowledge Organization Systems for Digital Humanities”, Queerlit is highlighted as an example where input from subject and information specialists is crucial to ensure information discovery. Queerlit has also developed a module within the course “Social Justice in the Digital Humanities” on transformative metadata.
The project aimed to achieve three key objectives:
1. Create a bibliography for Swedish lgbtqi literature in Libris.
2. Develop a thesaurus for improved subject description of this literature.
3. Design an external interface for the bibliography that offers enhanced search capabilities.
Queerlit provides new opportunities to identify literature that either represents lgbtqi experiences or has been interpreted as queer in research or queer communities. The Queerlit bibliography includes fictional works that portray same-sex sexual practices, emotions, and/or identities, as well as depictions of transgressions of binary gender norms through body, clothing, and/or identity. The project established inclusion criteria available on Queerlit’s website, applying these systematically. The bibliography incorporates works that explicitly address LGBTQI themes as well as those where such meanings are evident in context. This “context criterion” captures the history of lgbtqi literature, where such themes were often implicitly rather than explicitly represented.
The QLIT thesaurus was developed based on the largest international controlled vocabulary for LGBTQ+ materials, the Homosaurus. Terms were translated and adapted to Swedish contexts, and mapped to Homosaurus, Swedish Subject Headings (SAO), Children's Subject Headings, and Library of Congress Subject Headings (LCSH). These terms were used to enhance the subject descriptions and searchability of works included in the Queerlit bibliography.
The Queerlit interface provides better search capabilities than Libris web search. It was launched in February 2023 and developed iteratively through workshops and feedback from the indexing team. The interface development and linked thesaurus were created in dialogue with the National Library of Sweden (KB), addressing integration of thesaurus data into Libris and adapting the Libris search API for Queerlit’s needs. The interface enables term-based searches and visualizes the hierarchical structure of the thesaurus, showing relationships between broader and narrower terms.
PROJECT OUTCOMES
The Queerlit bibliography currently contains 2,375 titles in Libris, surpassing the initial goal of 1,000 titles. It spans works from the 6th century to the present. Materials were identified through prior research, subject term searches in SAO and the Children's Subject Headings, full-text searches in the Swedish Literature Bank and Project Runeberg, student theses, tips from the public and librarians, social media, community workshops, and monitoring of new publications. Literature published before 2000 was accompanied by descriptive texts justifying its inclusion.
By the end of the project, the thesaurus contained 913 terms, with 779 matching Homosaurus, 208 matching SAO, and 61 matching Children's Subject Headings. However, 120 terms lacked equivalents in any of these systems. Each term includes a scope note defining its use. A list of 81 terms deemed broad enough for SAO was submitted to KB and integrated into SAO.
Indexing in the bibliography diverged slightly from traditional fiction indexing. QLIT terms were pre-developed, with new terms added iteratively during indexing. A “user warrant” approach prioritized terms likely sought by users, and exhaustive indexing was applied, deviating purposely from Libris norms. QLIT is an openly available resource, with its subject terms linked in Libris. Indexing guidelines were developed in collaboration with KB and KvinnSam.
User testing with librarians and other users was utilized in the interface development. Research on improving interfaces to leverage underlying knowledge organization informed the visualization of thesaurus term relationships.
RESEARCH INITIATED WITH THE INFRASTRUCTURE
The project established an advisory board with subject experts and a Libris representative. This group supported strategic decisions regarding inclusion criteria, the interface, and thesaurus development. Collaborative proposals with national and international researchers were submitted to funding bodies like VR and Chanse Hera, although not funded. These application collaborations contributed to establishing contacts with scholars focused on minority information access in European libraries and digital cultural heritage infrastructures useful for further research collaborations. Queerlit also cooperated with the Nordic Network for Queer Archives and Activities (NNAQH) and the Dawit Isaak Library on a database of banned books.
The search needs of users were examined in two studies in 2022 and 2023, one directed at general users and one at librarians. The results demonstrated the need for an infrastructure like Queerlit in both groups. Users lacked specific terms and found the search interfaces to be user-unfriendly. Professional subject index terms assigned by information specialists using the QLIT thesaurus have also been compared to the performance of automated solutions like ChatGPT and Annif, and QLIT has been compared to social tags.
CHALLENGES
Queerlit has served as a pilot project to test the new possibilities in Libris XL, which supports linked open data. QLIT is the first external subject heading system that can be used in Libris, and this work has tested both the possibilities of making external data available for import into Libris and using the Libris API to display the bibliography through an external search interface. It was not possible to fully implement all desired search functions due to limitations in Libris’s software platform. However, it should be emphasized that through the dialogue with Queerlit, the need to modify LibrisXL’s API was identified, which may also benefit other projects.
MANAGEMENT AND ACCESSIBILITY
After the project's conclusion, KvinnSam, according to an agreement between the Gothenburg University Library and the fund manager, is responsible for indexing new titles in the database based on suggestions from a working group. The working group consists of four members from the Expert Network for LGBTQ+ Issues within the Swedish Library Association, including a former member of Queerlit. The working group also meets with a representative from KvinnSam twice per year.
An editorial team for the QLIT thesaurus has been established, consisting of four members from Queerlit, one of whom is employed at KvinnSam, as well as a representative from SAO. Indexing guidelines have been developed and are openly published on the project’s website, as well as linked from Libris. GRIDH, Gothenburg Research Infrastructure for Digital Humanities, is responsible for ensuring that the external interface functions and is updated. The code for the interface is openly available on GitHub.
ITERNATIONAL COLLABORATIONS
The chair of the Homosaurus editorial board has acted as an advisor to the project and as a contact person for interoperability between QLIT and Homosaurus. Queerlit has submitted a list of false matches between LCSH and Homosaurus identified during the work with QLIT, as well as suggestions for corrections to the hierarchical term structure. Thus, Queerlit has contributed to improve the quality of Homosaurus. Work to translate Homosaurus into various languages has been initiated, and in 2024, the translation of terms in Homosaurus into Swedish began in collaboration with Queerlit. As of 2025, a project member has joined the permanent editorial board of Homosaurus.
The Scientific and Technical Advisory Council (STAC) at the International Society for Knowledge Organization (ISKO) was established as a result of the Queerlit project to create international guidelines for end-user interfaces based on subject metadata available in many library catalogs but rarely utilized at the interface level. Search functionalities proposed by ISKO have been tested and implemented to the extent possible in the Queerlit database and will be used as ‘best practice’ examples for various functionalities.
Queerlit has also given rise to research on subject metadata in search systems in collaboration with the International Society for Knowledge Organization (ISKO), and discussions have begun with the International Federation of Library Associations and Institutions (IFLA) regarding international guidelines for subject indexing of lgbtqi resources.
The DARIAH-EU infrastructure for digital humanities has a platform for open educational resources, DARIAH Teach. In the course “Introduction to Knowledge Organization Systems for Digital Humanities”, Queerlit is highlighted as an example where input from subject and information specialists is crucial to ensure information discovery. Queerlit has also developed a module within the course “Social Justice in the Digital Humanities” on transformative metadata.