SwePub is Sweden’s national aggregated database of scholarly publications, harvesting publication metadata from more than 40 Swedish higher-education institutions (HEIs) and public research organisations into a single searchable service. It is run by the National Library of Sweden (Kungliga biblioteket, KB), and it functions as the country’s shared bibliographic layer for research output — the publication-level counterpart to the local research-information systems each Swedish university runs internally.
For research administrators working with Swedish institutions, or anyone assembling a publication list, compiling a bibliometric analysis, or checking a Swedish researcher’s output, SwePub is the closest thing Sweden has to a single point of reference, even though it is explicitly an aggregator rather than a system of record in its own right.
What SwePub is, and what it isn’t
SwePub does not hold original publication records. Each contributing institution maintains its own local institutional repository or research-information system — commonly built on platforms such as DiVA (a shared repository platform used by many Swedish and other Nordic universities) or a locally deployed Pure instance — and SwePub periodically harvests metadata from those local systems. This makes SwePub an aggregation layer, in the same functional category as other national publication-metadata hubs, rather than a mandatory government reporting registry with its own primary data-entry workflow.
That distinction matters in practice: a correction to a publication record needs to be made at the contributing institution’s own repository, not directly in SwePub, and then re-harvested on SwePub’s next update cycle.
Governance: the National Library of Sweden
SwePub is operated by the National Library of Sweden (KB), which under Sweden’s research policy framework holds national responsibility for coordinating work on open access to scientific publications (a division of labour set out in Swedish government research bills, with the Swedish Research Council, Vetenskapsrådet, correspondingly responsible for national coordination on open access to research data). Running SwePub sits naturally within that mandate: a national aggregation of publication metadata is infrastructure that supports open-access monitoring, discovery, and bibliometric work across the whole higher-education sector, not something any single university could usefully build alone.
SwePub’s own codebase is published openly, and KB documents the service’s data model, harvesting behaviour, and access methods on its website rather than treating them as internal implementation detail — consistent with running SwePub as shared public infrastructure rather than a proprietary product.
How the aggregation works
SwePub harvests structured metadata — not full text by default, though many records link out to or embed a full-text copy where the contributing repository provides one — from participating institutions’ repositories using standard interoperability protocols:
- OAI-PMH (Open Archives Initiative Protocol for Metadata Harvesting), the long-established protocol most institutional repositories already expose for exactly this kind of periodic harvesting.
- SRU (Search/Retrieve via URL), for structured query-based retrieval.
- Open APIs returning metadata in JSON, including a Swepub-specific BIBFRAME-based JSON representation and a MODS-based XML format, plus a lightweight HTTP query API (Xsearch).
- Bulk data dumps, for users who want the full corpus rather than incremental or query-based access.
Because harvesting draws on whatever metadata quality and completeness each contributing institution’s local system provides, SwePub’s own documentation and tooling put real emphasis on data-quality validation and classification support for the institutions and librarians who feed it — it publishes separate services aimed specifically at bibliometricians and librarians for checking data status and subject classification, on top of the public search interface aimed at researchers and students.
What you’ll find in it
SwePub covers the standard range of scholarly output types Swedish institutions report: peer-reviewed journal articles, conference contributions, and doctoral and licentiate theses, along with other publication types contributing institutions choose to submit. Records can be filtered by publication type, author, subject, and contributing organisation, and records marked as available in full text can be read directly through SwePub rather than requiring a separate lookup.
Role in Swedish research administration
SwePub’s practical value for research administrators is less about any single mandatory reporting obligation (Sweden does not run a POL-on- or Cristin-style statutory national CRIS registry in the same way some other European countries do) and more about being the shared, searchable aggregate of Swedish research output that several downstream uses draw on:
- Bibliometric analysis. Because SwePub is the most complete cross-institutional corpus of Swedish scholarly output available in one place, it is a natural source for bibliometric studies of Swedish research — whether run by an individual institution’s library, a funder, or a national body — rather than requiring bibliometricians to query dozens of separate institutional repositories individually.
- CVs and publication lists for grant applications. Swedish researchers routinely draw their publication lists for funding applications and academic CVs from their institution’s local repository (DiVA or Pure), the same record that SwePub subsequently aggregates — so a clean, complete, correctly attributed local record is what ultimately determines how well that researcher’s output is represented both in their own applications and in any SwePub-derived analysis of their institution.
- Open-access monitoring. As the national aggregation point sitting under KB’s open-access coordination mandate, SwePub metadata (including licence and full-text-availability fields where reported) supports tracking open-access uptake across Swedish institutions over time.
- Institutional reporting and benchmarking. Universities and university colleges use their own SwePub-aggregated output, filterable by organisation, as a convenient cross-check against their internal repository counts and for comparing output volumes and types against peer institutions.
Research administrators should treat SwePub as a discovery and analysis layer sitting on top of institutional data, not as the authoritative record itself — the authoritative record for any single publication remains whichever institutional repository originally supplied it.
SwePub and Sweden’s other national research infrastructure
SwePub is a publication-metadata aggregator, not a funding or licensing body, and it should not be confused with the Bibsam Consortium — also coordinated by the National Library of Sweden — which negotiates national open-access and licensing agreements with publishers on behalf of Swedish institutions. The two sit under the same host organisation and both support Sweden’s open-access agenda, but they do different jobs: Bibsam negotiates the agreements that make publishing and reading open access possible and affordable; SwePub aggregates and exposes the metadata of what has actually been published as a result.
SwePub also does not issue persistent identifiers. It aggregates metadata that typically already carries identifiers assigned elsewhere — a DOI for the publication itself, an ORCID iD for an author, a ROR ID for an affiliated organisation — rather than minting its own. Swedish institutional repositories that feed SwePub increasingly capture these identifiers as part of the record, which is what allows SwePub search results to be cross-referenced reliably against international identifier infrastructure and against interoperability standards such as CERIF, the data model maintained by euroCRIS for exchanging research information across systems and borders.
Frequently asked questions
Is SwePub a mandatory reporting system for Swedish universities?
No. SwePub is an aggregator that harvests metadata from institutions’ own repositories (commonly DiVA or Pure); it is not itself a statutory registry with a direct reporting obligation in the way some other countries’ national CRIS systems work. Institutions do, however, have their own internal expectations and open-access policies governing what gets deposited in their local repository, which is what SwePub then aggregates.
Who runs SwePub?
The National Library of Sweden (Kungliga biblioteket, KB), which holds national responsibility for coordinating Swedish open-access-to-publications policy and infrastructure.
How is SwePub different from the Bibsam Consortium?
Both are coordinated by KB, but they do different jobs. Bibsam negotiates national licensing and open-access publishing agreements with academic publishers on behalf of Swedish institutions. SwePub aggregates and makes searchable the metadata of publications that Swedish institutions have actually produced. Bibsam is about the agreements that enable publishing and access; SwePub is about the resulting record of output.
Does SwePub host full text?
Not universally. Some records link directly to a full-text copy hosted by the contributing institution’s repository and are readable in SwePub; others provide metadata only, with the full text (if available at all) accessible through the originating institutional repository or publisher.
Can I access SwePub data programmatically?
Yes. SwePub metadata is freely available via OAI-PMH, SRU, open JSON/XML APIs (including a BIBFRAME-based format and a lightweight Xsearch HTTP API), and bulk data dumps, and its underlying code is published openly.
Does SwePub assign identifiers like ORCID or a DOI?
No. SwePub aggregates metadata that typically already carries identifiers assigned by other systems — DOIs for publications, ORCID iDs for authors, ROR IDs for organisations — rather than minting its own.
