What Open Journal Systems is
Open Journal Systems (OJS) is open-source software for managing and publishing scholarly journals, developed by the Public Knowledge Project (PKP), based at Simon Fraser University in Canada. It is built for one process that no generic CMS handles well: the editorial workflow of a peer-reviewed journal, from the moment an author submits a manuscript until the reviewed and copyedited article appears in an issue. According to usage data published by PKP, more than 58,000 active journals in about 156 countries used OJS to publish at least five articles in 2025.
OJS organizes every manuscript in four stages, each with its own files and discussion thread:
- Submission. The author uploads the manuscript and its metadata through a guided form, and the editor decides whether it moves forward.
- Review. Peer review and the author's revisions, with reviewer invitations, deadlines and automatic reminders.
- Copyediting. Language, style and format editing, with the author's approval.
- Production. Conversion into the final galleys, such as PDF and HTML, and scheduling in an issue.
Roles (journal manager, editors and section editors, reviewers, authors, copyeditors and layout editors) see only the stages and files they need. The result is a process that is visible and auditable: who has each manuscript, in which stage, and since when. OJS does not make peer review faster or replace the editorial board's judgment, but it replaces the email threads and spreadsheets that most journals still run on. It is not a repository either: theses and reports belong in an institutional repository, and both systems share identifiers such as DOIs.
How BiblioLabTech implements OJS
Scope and editorial policies first
We start by confirming whether the journal is new or already publishing elsewhere, how many back issues exist and in what state, and how many journals the installation will host. We also check that the journal's policies are written down: review model, deadlines, rejection criteria and who holds each role. Configuring a workflow that has not been defined only moves the confusion into software.
Installation on a supported release
OJS is installed on the current long-term-support line. PKP designated OJS 3.5 as its long-term-support release in 2026, and BiblioLabTech does not install branches whose end of support is already in sight. The installation includes the journal's own domain or subdomain, HTTPS and a backup policy that covers both the database and the files, because losing either one leaves the system inconsistent.
Back issues and older installations
Migrating back issues is the step most often underestimated. Depending on what exists, we use structured import from CSV or XML, direct migration from another OJS installation, or Quick Submit when metadata only lives in PDFs. The first deliverable is an inventory of which issues exist, in what format and with how complete their metadata is; that inventory decides whether the migration takes two weeks or two months. Installations older than OJS 3.3 must upgrade to 3.3 before moving to the current series, as PKP requires, and we always work on a copy, verifying articles, galleys, users and DOIs before switching production.
Crossref DOIs
If your journal is a Crossref member with its own DOI prefix, we configure the Crossref plugin, choose the stage at which DOIs are assigned and set up automatic deposit, ideally before the first issue so identifiers never need to be reassigned. Registered DOIs keep resolving after a migration, and permanent redirects protect links published elsewhere. Crossref membership and per-DOI fees are paid directly to Crossref, not to BiblioLabTech.
Indexing readiness: what OJS can and cannot do
No software provider can guarantee indexing; each index decides on the journal's editorial merit. What OJS does is make the evidence visible and verifiable. During implementation we configure what the platform can sustain: published editorial and open access policies, peer-review traceability, explicit licenses per article, persistent identifiers, ORCID iDs for authors, complete metadata in several languages, OAI-PMH for harvesters and export to DOAJ. DOAJ, for example, expects the open access policy and the peer-review process to be visible on the journal's public site, which OJS shows when its configuration is complete. Latindex and the SciELO collections apply their own criteria, and several SciELO collections require prior DOAJ indexing. We also set up deposit in the PKP Preservation Network from the first issue, a safeguard several indexes value.
Hosting and support
OJS runs on Linux with PHP and a MySQL, MariaDB or PostgreSQL database. BiblioLabTech hosts it on managed infrastructure with HTTPS, backups, monitoring and security updates on the long-term-support release, or installs it on your own servers with documentation. We size storage for the original submission files, which usually weigh more than the published galleys. Annual support covers incidents, server and backup maintenance when we host, planned upgrades, functional help for the editorial team and periodic checks of the DOI and indexing configuration. Our guide to OJS setup and hosting walks through each step.
Editorial teams are trained by role: editors and section editors, reviewers, copyeditors and layout editors, authors, and site administration for the library or university press. Training runs on your real journal with a test manuscript that goes through the full workflow, because an editorial process is learned by doing it.
One installation for your whole portfolio
OJS is multi-journal. A university press or library can publish every journal of the institution from a single installation, each with its own editorial team, policies, sections, email templates and visual identity, under one central administration. That lowers the cost per journal and simplifies upgrades. OJS metadata can also be harvested by your DSpace repository and by your library's metasearch engine, and users can authenticate against your institutional directory.
What drives the cost of an OJS project
OJS has no license fee. The cost depends on the number of journals, the size and quality of the back file, the integrations and who hosts the system. These are reference ranges from BiblioLabTech's projects, covering implementation and migration in the first year:
| Scenario | First year (USD) |
|---|---|
| New journal, no back issues | 1,500 to 3,000 |
| Existing journal, moderate back file | 3,000 to 6,000 |
| Long back file with incomplete metadata | 6,000 to 10,000 |
| Several journals in one installation | Per journal, sharing the base infrastructure |
The ranges exclude Crossref fees and separately contracted hosting. Producing articles in JATS XML is an editorial service distinct from implementation and moves the price. After the first year, the recurring cost is usually a fraction of the implementation: hosting if managed, security updates on the supported release and support.
In effective work, a new journal without back issues needs three to five weeks; an existing journal with structured metadata, five to eight weeks; a long back file with incomplete metadata, eight to fourteen weeks. Calendar time depends on how often the editorial board can meet to take decisions.
BiblioLabTech works remotely with journals and university presses in any country, from Colombia (UTC−5), in English, Spanish, Portuguese and French, with the same team that runs Koha and DSpace in production for universities. Tell us about your journal and we will reply with a scoped proposal based on your real back file.
OJS is free software developed by the Public Knowledge Project at pkp.sfu.ca.