# KNOWN_LIMITATIONS.md # Known Limitations ## Overview The OpenAlex Publication Discovery and Repository Enrichment Toolkit has been developed to support repository workflows using openly available metadata from OpenAlex. The toolkit has been validated against live University of Kent repository data and is designed to minimise manual intervention while recognising that some publication records require human review. The following limitations should be considered during normal operation. --- # Metadata availability OpenAlex can only provide metadata supplied by publishers, repositories and other upstream metadata providers. As a result: * recently published outputs may initially contain incomplete metadata; * volume, issue and page numbers may become available some time after publication; * DOI information may appear after a repository record has already been created. The metadata enrichment report has been designed to identify these situations. --- # Publication lifecycle A scholarly work may legitimately exist in multiple forms, including: * preprint * accepted manuscript * published article These records may have different DOIs or no DOI at all. The toolkit treats these conservatively and routes potential publication lifecycle matches to **manual review**. --- # Records without DOI Some legitimate scholarly outputs do not have a DOI. Examples include: * institutional repository deposits * conference proceedings * reports * publications from smaller journals * teaching and professional outputs Where possible the toolkit attempts title matching, but records without DOI may require additional staff review. --- # Repository snapshot Duplicate detection is performed against a periodically exported repository snapshot rather than the live repository. This approach provides: * improved performance; * reproducible results; * reduced technical complexity. However, records added to the repository after the export was created will not be recognised until the next toolkit run. --- # Metadata enrichment The enrichment report identifies metadata available through OpenAlex that appears to be absent from the repository export. Suggested updates should always be reviewed by repository staff before being applied. The toolkit recommends potential improvements but does not modify repository records automatically. --- # Manual review The toolkit deliberately adopts a conservative approach. Manual review may be required where: * publication lifecycle events are detected; * DOI information differs; * title matching is ambiguous; * metadata conflicts cannot be resolved automatically. This reduces the risk of incorrect automated decisions. --- # Character encoding Metadata obtained from different sources may occasionally contain differences in: * punctuation; * quotation marks; * dash characters; * accented characters; * HTML entities; * character encoding. The toolkit performs title normalisation to reduce unnecessary manual review, but occasional differences may still occur. --- # OpenAlex coverage OpenAlex provides broad coverage of scholarly outputs, including: * journal articles * books * book chapters * datasets * reports * other scholarly works No discovery service is completely comprehensive and coverage depends upon the availability of metadata from upstream providers. The toolkit should therefore be regarded as a significant enhancement to repository discovery rather than a complete replacement for institutional quality assurance. --- # Future development The toolkit has been designed to evolve as repository workflows and OpenAlex continue to develop. Potential future improvements include: * additional publication types; * enhanced metadata comparison; * improved title normalisation; * expanded reporting; * support for additional repository platforms. These developments are intended to improve automation while retaining appropriate opportunities for repository staff to exercise professional judgement.