The Journalist's Information Challenge
An investigative journalist working on a long-form story may spend months or years accumulating information:
- Sources: Whistleblowers, insiders, experts, officials — each with varying levels of confidentiality and attribution constraints.
- Documents: Leaked materials, public records, financial filings, correspondence, meeting notes, interview transcripts.
- Entities: Companies, government agencies, non-profits, and the individuals who run them.
- Events: Meetings, transactions, decisions, incidents — placed on a timeline to reveal patterns and causation.
- Relationships: Who knows whom, who works for whom, which companies are connected through shared directors or owners, which entities have financial relationships.
This is not a collection of Word documents in a folder. It is a structured dataset with explicit connections. Managing it well is the difference between publishing a story that holds up under scrutiny and missing the connections that make the story significant.
The Source Protection Imperative
Source protection is the defining requirement for any tool used in investigative journalism. It is not sufficient for the software to have "privacy settings" or "access controls" if the underlying architecture exposes data to a third party. Cloud-based tools — note apps, shared drives, web-based research organizers — all share one fatal characteristic for journalism: the provider can access the data. Whether through automated scanning, response to legal process, or security breach, the data is not fully under the journalist's control.
A locally installed research platform eliminates this vector. The database and documents live on the journalist's own computer or a LAN server. There is no provider who can be subpoenaed, compromised, or pressured. The software architecture itself enforces source protection — not as a feature, but as a default property of local installation.
Essential Features for Journalistic Research
Entity and Relationship Mapping
The core of investigative research is understanding how people, companies, and organizations connect. A research CRM must model these connections explicitly: Person A is a director of Company B, which is a supplier to Government Agency C, which is run by Person D, who used to work at Company E alongside Person A. The relationship graph makes these chains visible.
Source Management with Attribution Levels
Not all information has the same attribution constraints. Some sources are on the record; others are on background; still others are confidential with strict conditions. The system must track the attribution level for each piece of information and each document, so the journalist can assess what can be published and how.
Document Organization by Entity and Event
Documents should be linked to the entities and events they pertain to, not just stored in a folder. When the journalist needs everything related to a particular company or a particular meeting, the relevant documents should be retrievable through those associations, not through memory of the folder structure.
Timeline Construction
Chronology is central to investigative storytelling. Events linked to entities and supported by documents create a verifiable timeline. Gaps in the timeline become visible and can be pursued as leads.
Offline Capability
Journalists often work in environments where internet connectivity is unreliable, monitored, or intentionally disabled. The research database must be fully functional without any network connection — a natural property of local-first software.
ONS Data Terminal for Journalistic Research
ONS Data Terminal provides entity management, document linking, relationship graph visualization, and timeline features — all running locally on the journalist's machine or a dedicated research server. Data is stored in a local PostgreSQL database and on the local filesystem. There is no cloud component, no telemetry, and no external dependency. The journalist's research data is as private as the computer it runs on.