JabRef

1. Overview

Reference manager built directly on BibTeX and BibLaTeX files.

Category: Research & Education.

2. Version and scope assessed

Version assessed: Source commit 1e2e9a089986e317cd7f82f4c0114b8950164925; BibTeX grouping fixture and group/file-field serializers inspected.

4. Principle profile

5. Fundamental requirements

6. Evidence and verification

• E1: Resolved: the retained BibTeX fixture exposes two bibliographic records and a jabref-meta grouping comment with the root and four named groups. Group depths identify Papers with 2024 and Machine Learning children. This is text inspection, not a complete BibTeX parser or group-membership round trip. (source), confidence high
• E2: The application and command line support exporting libraries and selected entries to BibTeX and other formats, enabling a practical exit path. (source), confidence high
• E3: Resolved: FileFieldWriter encodes description, link, file type, and optional source URL with colon/semicolon delimiters and backslash escaping. GroupSerializer records depth and recursively serializes children. The representations are available, while linked-file existence, migration, and full restoration remain unverified. (source), confidence medium
• E4: The core library is stored in local files and can be used without a hosted account, while optional web fetchers are separate conveniences. (source), confidence medium
• E5: Bibliography management is user-directed, but automated metadata fetching and citation consequences were not tested for review or correction safeguards. (source), confidence low
• sovereignty: A positive Sovereignty finding requires established E1, E2 and E4 in the declared scope. The existing access/exit/custody record for JabRef does not establish all three minimums. The prior positive finding is withdrawn; recorded storage and portability strengths remain facts, not a verified failure. (source), confidence medium
• longevity: E3 is not established for JabRef's existing essential-record scope. The previous Longevity finding is replaced by Not assessed, not a verified failure or a positive placeholder. (source), confidence medium
• locality: The primary library workflow is local-file-based and does not require a central service. (source), confidence medium
• transparency: The positive control observations above do not yet establish whole-scope E5. The methodology requires that minimum before a positive Transparency and Control finding; the previous positive finding is therefore withdrawn to Not assessed, not changed to Does not meet. (source), confidence medium
• E3: The BibTeX sample exposes records/group hierarchy and serializers document escaped file fields and recursive group depth. Complete memberships, custom fields/types and existing linked-file relocation have not been reconstructed with an independent reader. Bibliographic text readability alone does not cover the original research profile. (source), confidence medium
• E5: User-directed editing and inspectable BibTeX are not complete safeguards for automatic metadata replacement, file rename/move/delete, merge operations and citation changes. The collected grouping/file-field evidence does not sufficiently establish authorization, visible diffs and correction for those operations. (source), confidence medium
• E2: The prior E2 Pass does not establish independent-workflow preservation of essential group memberships, custom metadata and linked files. Exposing serializer formats or a command-line export is not that migration comparison. (source), confidence medium

7. Migration and recovery

Native storage format: BibTeX and BibLaTeX plain-text files (.bib).

Synchronization or runtime model: Local files; versioning and sync left to external tools.

Example export or inspection command (not verified against any particular version or data set): grep -c '^@' library.bib

8. Architectural strengths

Established strengths and their scope are recorded in the evidence and assessment findings on this page.

9. Architectural limitations

• Group definitions use JabRef metadata comments. File fields store escaped links and optional source URLs, not the linked file bytes.

10. Assessment findings and remaining work

Assessment update, 23 October 2026:

• Resolved: A real BibTeX fixture exposes the root and four named groups, including two nested children under Papers.
• Resolved: The source specifies associated-file encoding, including escaping and the optional source URL.
• Open: Entry memberships, custom types/fields, other readers’ handling of metadata comments, linked-file relocation, and cross-version recovery. E5 needs metadata-fetch/merge/file-operation review, visible changes and rollback or recourse for the configured library.

Assessment: The selected library’s group definitions and the associated-file representation are independently inspectable. These are resolved storage questions; complete library and attachment recovery remains Not Verified. The overall determination remains Insufficient evidence.

Human-readable criteria: Can a user preserve records, structure, media, and provenance, restore them independently, and move to another tool without losing meaning?

Rating reconciliation: 24 October 2026

E3 and E5 remain Not Verified for the existing declared scope. Longevity and Transparency and Control are Not assessed pending their required minimums. Sovereignty is also Not assessed because the required access/exit/custody minimums are not all established. The prior E2 Pass is withdrawn; its essential-record migration comparison remains incomplete. This is a targeted evidence decision, not a finding that the product fails. Existing E1, E4 and Locality findings were not re-certified.

Remaining: Entry memberships, custom types/fields, other readers’ handling of metadata comments, linked-file relocation, and cross-version recovery. E5 needs metadata-fetch/merge/file-operation review, visible changes and rollback or recourse for the configured library.

Next without installing locally: Use a populated synthetic BibTeX library and actual linked-file tree with expected groups/memberships; independently parse it and seek version-matched automated-metadata, merge and file-operation control evidence.

The individual decision and provenance are retained in evidence/rating-reconciliation/batch-5/. Overall: Insufficient evidence.

Individually reconciled in rating batch 5 on 24 October 2026. E3/E5 remain Not Verified for explicit missing coverage; dependent unsupported findings corrected. Previously resolved questions are retained, not counted again. E1/E4/Locality were not re-certified by this targeted review. The assessment procedure is described in the methodology.