VIVO is the only widely-adopted current research information system (CRIS) that is open source, community-governed, and built on a semantic-web/Linked Open Data model rather than a proprietary relational database. Research offices evaluating a CRIS deployment usually put it on the shortlist next to commercial platforms such as Symplectic Elements, Pure, and Converis — but VIVO’s technical architecture, governance model, and total cost of ownership are different enough from those products that a straight feature-by-feature comparison table understates what actually distinguishes it. This guide covers VIVO on its own terms: what it is, how it’s built, who governs it, what running it in-house actually takes, and when it is — and isn’t — the right choice.
For a side-by-side feature comparison against the commercial platforms, see Pure vs Symplectic Elements vs VIVO vs DSpace-CRIS and, for the vendors that don’t include VIVO, Symplectic Elements vs. Pure vs. Converis: A CRIS Procurement Scorecard.
What VIVO is
VIVO is open-source software and an accompanying ontology for representing scholarship as linked data. Rather than storing researcher, publication, grant, and organizational records as rows in a relational database — the model every major commercial CRIS uses — VIVO expresses that same information as RDF triples in a semantic-web triple store, queryable via SPARQL. The VIVO Ontology is an OWL ontology that imports and extends established vocabularies including BIBO (bibliographic data) and FOAF (people and organizations), so that a VIVO instance’s data is, by construction, expressed in terms other linked-data systems can consume without a custom export format.
Practically, this means a VIVO installation is a Java web application (built on the Apache Jena semantic-web framework) sitting in front of a triple store, with a public-facing researcher-profile interface, an editing interface for data stewards, and a SPARQL endpoint for programmatic and federated querying. Participating institutions’ VIVO data can also be harvested into cross-institutional search and linked-data graphs — a design goal from the outset, not a bolt-on integration.
VIVO is released under the BSD open-source license: the source code is free to use, modify, and redistribute, with no per-seat or per-record vendor licensing fee.
Origins and governance
VIVO began in 2003 as a Cornell University Library project, led by Jon Corson-Rikert at Mann Library, to connect researchers and resources across Cornell’s life-sciences departments. In September 2009 the NIH’s National Center for Research Resources awarded a two-year, $12.2 million grant — formally “VIVO: Enabling National Networking of Scientists” — to a consortium of Cornell University, the University of Florida, and Indiana University Bloomington, together with implementation partners, to build “VIVOweb”: a multi-institutional version of VIVO connecting biomedical researchers, initially across seven sites.
VIVO became a DuraSpace-supported open-source project in 2013, moving from a single-institution effort to a community-governed one with a member-supported foundation, elected leadership bodies, and open contribution processes. On July 1, 2019, DuraSpace completed its merger with LYRASIS; VIVO’s stewardship, along with DSpace and Fedora, transferred to LYRASIS’s DuraSpace Community Supported Programs division, which remains VIVO’s institutional home today. This governance history is the practical difference underneath the “open source” label: VIVO is not maintained by a single vendor that can be acquired, discontinue the product, or unilaterally change pricing — development priorities are set through a member community and public roadmap process, the trade-off being that a research office has less of a single point of contractual accountability than it would with a commercial vendor’s support agreement.
How VIVO’s architecture differs from Pure, Symplectic Elements, and Converis
Pure (Elsevier), Symplectic Elements (Digital Science), and Converis (Clarivate) are commercial, closed-source CRIS products built on proprietary relational data models, licensed to institutions on a subscription basis, and typically deployed as vendor-hosted SaaS. Their competitive differentiation is largely about breadth of automated data harvesting (from Scopus, Web of Science, ORCID, and similar sources), reporting/dashboard tooling, and vendor-managed integration maintenance — the trade-off institutions accept for not having to run the underlying infrastructure themselves.
VIVO’s differentiation is structural rather than feature-by-feature:
- Data model. A relational CRIS models research information as a fixed, vendor-defined schema; extending it to capture an institution-specific entity type or relationship usually means requesting a vendor customization. VIVO’s RDF/ontology model is natively extensible — an institution can add classes and properties to the ontology itself, and the same data can be published as linked open data without a separate export step.
- Licensing and cost structure. Commercial platforms carry recurring vendor licensing fees in exchange for hosting, support, and managed upgrades. VIVO has no license fee, but the institution (or a contracted hosting partner) absorbs the infrastructure, development, and maintenance cost directly — see the deployment section below.
- Governance and roadmap control. Commercial roadmaps are set by the vendor, informed by customer advisory input. VIVO’s roadmap is set through its LYRASIS-supported member community and open development process; institutions with development capacity can contribute code directly rather than filing a feature request.
- Interoperability posture. Commercial CRIS platforms increasingly support CERIF-based exchange and standard PID integrations as vendor features. VIVO’s linked-data model was built around interoperability with the broader semantic web from the start, which is why it interoperates naturally with initiatives like CERIF-based CRIS interoperability efforts that also involve commercial platforms such as Pure.
Documented adopters and the VIVO community
VIVO’s user base spans universities, federal agencies, and international projects. Institutions and organizations documented as VIVO adopters or founding implementation partners in primary sources include Cornell University (the originating institution, whose libraries continue to document VIVO-based scholarly profile and publications-reporting tools), the University of Florida and Indiana University Bloomington (co-recipients of the 2009 NIH VIVOweb grant), the U.S. Department of Agriculture, the U.S. Environmental Protection Agency, the American Psychological Association, and the Australian National Data Service’s VIVO project. LYRASIS materials also describe institutions in the NIH Clinical and Translational Science Award (CTSA) Consortium among VIVO data producers, and vendors including Symplectic and Clarivate have built data-exchange connectors to VIVO instances — evidence that VIVO and commercial CRIS platforms are as often complementary as competitive in practice.
Rather than a single vendor-run customer list, VIVO’s community is organized through LYRASIS membership, open community groups (including a dedicated W3C VIVO Community Group), and public code repositories — a structure closer to an open-source software foundation than a product user base. Because institutional deployments come and go and LYRASIS’s own published adopter figures change over time, verify current membership and instance counts directly with LYRASIS or the VIVO project (vivo.lyrasis.org) rather than treating any fixed number as current.
What it takes to deploy and maintain VIVO in-house
The trade-off research offices actually weigh when VIVO is on the shortlist is not a feature checklist — it’s who runs the system. A commercial CRIS is typically vendor-hosted: the institution configures it and manages data governance, but the vendor operates the servers, applies upgrades, and maintains third-party data-source integrations under a support contract. VIVO is self-hosted software; running it requires:
- Server infrastructure to host the Java application and triple store, sized and secured to the same standard as any other institutional research-data system, and either operated internally or contracted to a hosting partner.
- Development/technical staff comfortable with Java, RDF/SPARQL, and ontology concepts to install, configure, customize the data model, and troubleshoot the application — a materially different skill set than administering a vendor SaaS console.
- Ontology and data-source customization work: mapping local data sources (HR systems, grants systems, publication feeds) into the VIVO ontology is a local implementation task rather than a vendor-delivered connector, though the open community has published reusable ingest tooling and mapping patterns that reduce — without eliminating — this work.
- Ongoing upgrade and security-patch responsibility, which in a vendor SaaS model is handled by the vendor as part of the subscription, and in VIVO’s model falls to the operating institution or its hosting partner.
The reward for that investment is full control over the data model, no recurring license fee, no vendor lock-in, and a data architecture that treats interoperability and linked-data publication as native rather than as an add-on. The cost is that VIVO substitutes an ongoing internal (or contracted) technical-operations commitment for a vendor subscription fee — a classic open-source build-versus-buy trade-off that research offices should size against their own IT capacity, not just their budget line, before committing.
When VIVO is — and isn’t — the right fit
VIVO tends to be the stronger fit when an institution already has, or is willing to build, in-house development capacity for research-information systems; wants full control over the data model and the ability to publish research information as linked open data; needs to integrate with semantic-web or linked-data infrastructure beyond what commercial CRIS connectors expose; or wants to avoid recurring per-seat vendor licensing in favor of investing that budget in internal staff and infrastructure. It also suits institutions and consortia (federal agencies, multi-institution research networks) that need a shared, non-proprietary data model for cross-organizational research-information exchange.
A commercial platform such as Pure, Symplectic Elements, or Converis tends to be the stronger fit when a research office wants vendor-managed hosting and upgrades, broad out-of-the-box harvesting from subscription databases like Scopus and Web of Science, dedicated vendor support under a contract, and does not have (or does not want to build) internal capacity for ongoing systems development. See the full CRIS vendor comparison for how VIVO stacks up feature-by-feature against these platforms, and the Symplectic Elements vs. Pure vs. Converis procurement scorecard for a deeper look at the commercial-only field.
Frequently asked questions
Is VIVO free to use?
VIVO’s software is free and open source under the BSD license — there is no vendor license fee. The institution still bears the cost of hosting, technical staff, and implementation work, either in-house or through a contracted hosting/implementation partner, so “free” refers to licensing, not total cost of ownership.
What is the VIVO Ontology?
It is an OWL ontology, maintained by the VIVO community, that defines the classes and relationships used to describe researchers, organizations, publications, grants, and courses as RDF data. It imports established vocabularies such as BIBO and FOAF and is designed to be locally extensible.
Who currently maintains VIVO?
VIVO is maintained as an open-source project under LYRASIS’s DuraSpace Community Supported Programs division, following LYRASIS’s 2019 merger with DuraSpace, which had stewarded VIVO since 2013. Development is community-driven, with governance and roadmap input from LYRASIS member institutions.
Can VIVO replace a commercial CRIS entirely?
For institutions with the technical capacity to run and customize it, yes — VIVO covers the same core research-information functions (researcher profiles, publication and grant records, organizational structure) as a commercial CRIS. The decision usually turns on whether the institution would rather invest in internal systems capacity or in a vendor subscription, not on a functionality gap.
Does VIVO integrate with identifiers like ORCID and DOIs?
VIVO’s RDF-based data model is designed to carry persistent identifiers such as ORCID iDs and DOIs as first-class linked-data properties on person and publication records, consistent with its broader design goal of interoperating with the wider identifier and linked-data ecosystem rather than treating identifiers as opaque reference strings.
See also: VIVO (dictionary entry), CRIS (Current Research Information System), Elements (Symplectic), and the CRIS & Identifiers pillar page for the broader research-information-management landscape.
