How Nodus can support research in the humanities

Humanities research often involves working with materials that do not fit neatly into a single research tool. A historian may need to move between books, archival documents, photographs, interviews, notes and databases. A literary scholar may work simultaneously with editions, annotations, secondary literature and a growing network of concepts. A researcher in genealogy or prosopography may need to keep track of people, places, relationships and uncertain evidence across many different sources.

Nodus is designed around this kind of work. Rather than treating research as a sequence of separate tasks, it provides a workspace where sources, documents, notes, ideas and structured information can remain connected. The application is local-first, so the material in a vault is stored on the researcher's own machine. AI can be used with an external provider or with compatible local models, while the underlying research material remains under the researcher's control.

This approach is particularly relevant to the humanities because the relationship between an interpretation and its source often matters as much as the interpretation itself. In Nodus, findings and ideas can remain connected to the passage, page or other piece of evidence from which they were derived. This makes the application useful not simply for storing research material, but for keeping track of how an argument develops from that material.

Working from a research library rather than a collection of disconnected documents

The Academic vault is the main starting point for humanities researchers. It can contain a research library built directly in Nodus or work with an existing Zotero collection. Sources can then become part of a wider network of claims, findings, methods, notes and ideas rather than remaining isolated records in a bibliography.

A practical workflow might begin with a literature review. A researcher imports a collection of papers or connects a Zotero library and reads the material inside Nodus. Important passages can then be connected to ideas or findings. As more sources are examined, relationships between them become easier to identify. Semantic search can help locate relevant material across the library, while the graph makes connections between sources and ideas visible.

This can also be useful when a research question develops gradually. Instead of creating a separate document every time a new idea appears, the researcher can connect the idea to the sources that support it and to other ideas that qualify or contradict it. Over time, the resulting structure can reveal areas where the literature is well covered, where different authors disagree and where a research question may still have little supporting evidence.

For researchers writing articles, chapters or theses, this provides a possible route from reading to writing. The material gathered during the literature review remains available when the argument is being developed, and citations can be traced back to the evidence behind them. Nodus also includes a writing environment and integrations with Word and LibreOffice, making it possible to move between research and writing without completely separating the two stages.

This kind of workflow can be particularly useful for historical research. A historian working with dozens or hundreds of books may want to identify how particular concepts, places or representations recur across different authors and periods. The important part is not simply finding another occurrence of a term, but being able to return to the source and understand the context in which it appeared.

Working with archival sources, people, testimony and structured historical data

Not all humanities research begins with a conventional academic library. Much of the work in history and related disciplines involves primary sources, photographs, manuscripts, interviews, catalogues and structured datasets. Nodus therefore includes several vault modes that address more specific forms of humanities research.

The Primary Sources vault is intended for working critically with archival documents and manuscripts. OCR and transcription can be used alongside the original document, allowing the researcher to distinguish what the source actually contains from what has been inferred from it. This distinction is important when working with handwritten material, imperfect scans or documents whose interpretation depends heavily on their original context.

The Testimony vault is designed around interviews and oral history. It supports local transcription, speaker separation and the recording of consent and access conditions. Passages can remain associated with clickable timestamps, which can make it easier to move from an analytical note back to the relevant part of an interview. This can be useful for oral historians, social historians and researchers working with recorded testimony.

The Genealogy vault provides another example of how the same underlying workspace can be adapted to a particular research problem. It allows researchers to reconstruct family histories and individual lives using primary sources, with a tree, timeline and map alongside GEDCOM import and export. Events can remain connected to the evidence supporting them. Although genealogy is often treated separately from academic historical research, many of its problems are familiar to historians, including uncertain identities, conflicting records, incomplete evidence and the need to distinguish documented information from inference.

Prosopography addresses a related problem at a larger scale. Instead of studying one individual or family, researchers can use it to examine groups and collective trajectories across scattered sources. Name variants and uncertain identities can be handled without necessarily hiding the uncertainty behind a single definitive answer. This can be useful for studies of political groups, intellectual networks, institutions, professional communities or other historical populations.

The Database vault can complement these more specialised modes when a project requires structured data. Researchers can build databases containing different field types, relationships, formulas, attachments and reusable filtered views. CSV import makes it possible to bring existing datasets into the workspace, while analysis tools can be used to examine the resulting data. For a historian, this might mean building a catalogue of photographs, a database of travellers, an inventory of archival documents or a dataset of places and events without having to maintain a completely separate research environment.

Connecting qualitative interpretation with quantitative and computational work

One of the difficulties in humanities research is that projects rarely remain purely qualitative or purely quantitative. A historian may begin by reading books and archival sources, then create a structured database to compare authors, dates, locations or themes. A literary scholar may combine close reading with a corpus of texts. A researcher working with photographs may need both detailed visual interpretation and structured information about thousands of images.

Nodus can accommodate these different forms of work within the same broader environment. A research library can contain the sources and notes used for interpretation, while a Database vault can hold structured information derived from those sources. The two approaches do not have to replace one another. They can answer different questions about the same research project.

AI can also be used at different points in this process. In an academic library, it can help extract ideas, identify relationships, explore contradictions or locate possible research gaps. In a database, AI columns can assist with classification while keeping the resulting values within the structure of the researcher's data. With local models through tools such as Ollama or LM Studio, researchers can also choose to keep AI processing offline.

The important point is that these capabilities are intended to assist research rather than replace the researcher's judgement. A generated finding still needs to be checked against its source. A possible relationship between two documents is not automatically an established relationship. A classification produced by an AI model is not necessarily a reliable historical category. Keeping evidence attached to findings helps preserve this distinction and makes it easier to return to the material when an interpretation needs to be reconsidered.

This becomes particularly useful in projects involving large collections. Consider a researcher working with several thousand photographs from a historical archive. The project might involve a Database vault containing structured information about each photograph, a Primary Sources workflow for examining the original material, notes recording interpretations and an Academic vault containing the secondary literature. AI could assist with some repetitive tasks, while the researcher retains responsibility for interpreting the photographs and verifying the resulting information.

The same principle applies to a historical corpus of books. A researcher might collect the bibliography in Zotero, examine the books and relevant passages in Nodus, record recurring ideas and connect them to particular sources, then use structured data to compare authors, dates, locations or themes. The result is not simply a collection of AI-generated summaries. It is a research environment in which the sources, interpretations and structured observations remain connected.

Choosing a Nodus workflow for different humanities projects

There is no single humanities research method, and Nodus is therefore organised around different vault modes rather than one universal project structure. Academic is the most general starting point for source-based research, particularly when the project revolves around a bibliography, documents, ideas and scholarly writing. Database is useful when the project depends heavily on structured information. Primary Sources is suited to archival documents and manuscripts, while Testimony provides a more specific environment for oral history and interviews.

Genealogy and Prosopography are useful when people and relationships form the central object of research. They can also complement broader historical projects. A study of an intellectual network, for example, could combine prosopographical information about individuals with an Academic library containing the relevant scholarship and primary sources. A project about a historical family could combine genealogical records with documents, photographs and contextual literature.

Teaching and Study are also part of the same application and can be relevant to humanities researchers who teach alongside their research. The Teaching vault can organise courses, materials, activities and assessment, while Study is designed around subjects, materials, notes, calendars and revision. This makes it possible to use the same application across different parts of academic work without requiring every activity to follow the same structure.

The most useful way to approach Nodus is therefore not to ask which single feature replaces an existing research tool. A better question is which parts of a particular workflow would benefit from being connected. A historian may continue using Zotero for bibliography management. An oral historian may continue working with specialist recording equipment. A researcher may still use a spreadsheet for a particular dataset. Nodus can sit alongside these practices while providing a common environment for the sources, evidence, ideas and structured information that make up the research project.

For humanities researchers, this is perhaps the central value of a research workspace such as Nodus. The application does not determine what an interpretation should be or turn historical evidence into conclusions automatically. Instead, it provides a way of keeping the different materials involved in research closer together. Sources can remain connected to claims, ideas can remain connected to evidence, structured data can remain part of the wider project, and AI can be used as an additional research aid without becoming the authority over the material.

That makes Nodus particularly suited to research projects in which the path from source to interpretation matters. Whether the material consists of books, archival documents, photographs, interviews, genealogical records or structured historical data, the aim is the same. Keep the evidence visible, keep the relationships between materials understandable and make it easier to return to the sources when the research develops.

The academic research guide explains the source-to-evidence workflow in detail. Researchers who already collect in Zotero can also compare the broader Zotero integration with the focused Nodus Zotero plugin.