EDUKATE SINGAPORE · GOLD STANDARD PRACTICAL CASE STUDIES · CASE 161
Two authors can share one name without sharing one identity or the same books.
A fictional library migrates seven book records and merges two unrelated writers named Asha Lim into one public author profile. This Gold Standard case shows why title import, text matching, authority identity and truthful student research require separate verification.
Both writers and all works are fictitious; the name is used only as a hypothetical example. Library of Congress guidance supplies established authority-control principles and does not identify any real author as involved in this event.
Chapter index
- 1. Seven books appear beneath one author, but there were two
- 2. A matching string is not a verified human identity
- 3. The catalogue’s arithmetic is neat but its biography is wrong
- 4. Library of Congress guidance gives an exact authority-control example
- 5. Subject matter is useful context but not a unique key
- 6. The public biography needs correction before it spreads
- 7. Privacy and authority records must be designed together
- 8. A better migration needs an accepted entity-level matching process
- 9. Retest with twins in name but not in identity
- 10. Practice: seven books, two writers, one mistaken profile
1. Seven books appear beneath one author, but there were two
A fictional neighbourhood library migrates its local catalogue to a new discovery platform. Four school mathematics books are written by one author using the name Asha Lim. Three collections of environmental poetry are written by a different person who also publishes as Asha Lim. The library’s source records contain distinct publication and creator information for each group.
During migration the matching tool compares the display-name field, sees the same string and combines all seven works beneath one creator profile. The search interface looks tidy: there is only one Asha Lim entry, and every book is reachable. A student preparing a presentation about the mathematics author then describes her as the author of three poetry collections she did not write.
The Gold Standard issue is not that name standardisation is inherently wrong. Libraries use authority control to connect genuine name variants and prevent fragmented discovery. Here, the system has merged two different identities because their names look identical. Successful matching of text did not establish successful identification of the person who created each work.
2. A matching string is not a verified human identity
The source data contain B1, four mathematics bibliographic records; B2, three poetry records; B3, available reliable creator identifiers and publisher evidence; B4, the old catalogue’s distinct creator links; B5, the migration’s name-matching rule; and B6, the public display after the merge. Each helps explain the difference between what the record says and who it identifies.
B5 may truthfully report that all seven works have the text ‘Asha Lim’ in their author-name field. It does not establish that they belong to one person. Two people can share a name, while one person can publish under several spellings or names. The matching programme needs a credible basis for choosing between those opposite cases.
The authority or creator record should stand for an identified entity under the library’s method rather than merely a repeated text label. When evidence is insufficient to determine whether two records refer to the same person, a responsible system should preserve uncertainty rather than force them together for visual neatness.
GS-R02 defines what counts as a valid identity match, while GS-R08 asks what evidence supports a claim that one human wrote all listed works.
3. The catalogue’s arithmetic is neat but its biography is wrong
The merged profile displays seven books. Four genuinely belong to the fictional mathematics writer and three to the fictional poet. Under the source data, four out of seven—approximately 57.1%—of the displayed entries belong to the first person, while three out of seven—approximately 42.9%—belong to the second.
Those percentages do not measure how often real library catalogues make this mistake. They describe the fictional profile’s composition under the stated source evidence. If a reader selects the first writer expecting all seven to be her works, the three poetry records create false creator attribution; if a reader expects the poet, the four mathematics books do the same.
All seven books remain real within the example, and the title and publication facts may be correct. The error lives in the relationship between a work and a creator identity. Accurate inventory counts and valid title records cannot compensate for that mistaken relationship when a teacher, researcher or student tries to learn who wrote what.
The Gold Standard recognises a fundamental information-system distinction: record existence, search retrieval and factual attribution are separate claims.
4. Library of Congress guidance gives an exact authority-control example
The Library of Congress explanation of MARC authority records describes authority control as using recognised identifiers and headings to bring together works by the same entity and provide cross-references from variant names. The purpose is consistent discovery, not treating every identical name as one individual.
The Program for Cooperative Cataloging’s Authority Control FAQs for Authors and Creators explicitly shows different authors who share names and explains why distinguishing biographical or professional information may be useful for identification. Its examples include multiple people known as Margaret Lee.
The Library of Congress MARC proposal on sources for names and titles contains an example in which the same name string, Henry Sargent, identifies different people under different source vocabularies. That demonstrates why an authority source and identifier can matter in addition to matching text.
These are established bibliographic practices, not proof that every matching problem should be solved by collecting more private details. The fictional Asha Lim examples merely apply the distinction to a learning-centre catalogue.
5. Subject matter is useful context but not a unique key
The library’s assistant assumes the mathematics writer and poet must be different because one group contains algebra and the other poetry. In this fictional story they indeed are different people, but subject difference alone is not conclusive evidence. A real writer can produce works in several disciplines and genres.
Relevant verification may use trustworthy publisher information, documented creator statements, existing identifiers, publication context and qualified cataloguing review. A professional should distinguish confirming facts from guessing identity from a person’s supposed interests, gender, ethnicity or writing style.
Another source might use a middle initial or a pen name. Such variants can genuinely represent one creator, so creating a new identity whenever a label differs would fragment works that belong together. Good authority control must avoid both false merging and unnecessary splitting.
The Gold Standard asks for traceable entity relationships and an honest unresolved state when evidence does not support a clear decision, rather than a simplistic ‘same string means same person’ rule.
6. The public biography needs correction before it spreads
The library preserves the original migration output, source bibliographic records and evidence supporting two creators. It restricts the misleading combined biographical statement until a qualified cataloguer can make and document an appropriate authority decision. The original records are not erased merely because the new interface made a mistake.
The student who used the profile should receive a clear correction: two distinct writers share the fictional displayed name, and the poetry titles should not be attributed to the mathematics author. That correction helps the learner revise the presentation without suggesting the student acted carelessly by trusting the library’s public record.
Management should not invent full dates of birth or private contact information to make the catalogue look authoritative. Actual identity differentiators should be chosen under recognised cataloguing standards, privacy policies and the evidence available. An uncertain author relationship can be flagged for review without accusing anyone of pseudonym misuse.
GS-R09 demands a prompt, bounded correction; GS-R03 identifies the person authorised to repair the metadata.
7. Privacy and authority records must be designed together
The PCC Position Statement on Personal Data in Name Authority Records, approved in 2023 explains that authority data help distinguish creators while creating responsibilities to consider the privacy impact of recording personal attributes. More information is not always better when identity can be established through a less intrusive method.
For the fictional catalogue, a stable creator identifier and reliable relationship evidence may be enough to separate the mathematics books from the poetry works without publishing home addresses or unrelated personal details. A publicly exposed authority record can be copied elsewhere, so even apparently minor identifiers deserve proportionate handling.
The Gold Standard protects both accurate attribution and personal dignity. A data migration should not turn privacy into collateral damage of a matching repair. GS-R07 governs the transfer from source catalogue to new platform; GS-R12 asks whether later changes preserve correct identity links.
8. A better migration needs an accepted entity-level matching process
The library revises its mapping so it uses available source authority identifiers and appropriate evidence, rather than treating a display name as a globally unique key. Where records conflict or lack sufficient metadata, the system queues a human-readable exception for competent review instead of silently merging identities.
Reviewers should check not only whether a reader can find all seven titles, but whether each title points to the right creator record and whether genuine name variants resolve appropriately. Search convenience is a meaningful outcome, but it does not justify fabricated author biographies.
The migration report can record how many records were imported, how many relationships were validated and which identities remain uncertain under the adopted method. A ‘100% titles migrated’ statistic is compatible with an unresolved attribution problem and should be labelled accordingly.
GS-R08 requires that the metric report the property actually tested, not a broader promise of identity accuracy.
9. Retest with twins in name but not in identity
The library creates a controlled test set containing two synthetic authors who share the same display name, plus one author whose name appears in two distinct spellings. The migration should preserve the two separate identities while bringing the verified variants of the third together under suitable authority evidence.
A deliberately weak matching rule is included to show how it wrongly merges the same-name authors and fragments the variant-name author. Cataloguers assess whether the output supports correct attribution in the public search interface, not merely whether all records pass a text-match completeness check.
The test also checks the corrected work-to-creator relationships in exports, citations and any public author profile. One corrected staff view does not establish that older copies in the discovery index and school worksheets have updated. The institution needs a controlled mechanism for propagating validated changes.
A passing exercise supports the selected rules and synthetic entities; actual cataloguing still requires professional judgement where identity evidence is incomplete.
10. Practice: seven books, two writers, one mistaken profile
Exercise. A fictional library merges four mathematics books by one Asha Lim with three poetry books by another Asha Lim into one seven-work author profile. A student attributes all seven to the first author. Write a bounded finding, the misleading aspect of the completion metric, the correction and one acceptance test.
A reasoned answer is: ‘All seven title records were imported, but the authority relation incorrectly conflated two distinct creators. Four entries belong to the mathematics author and three to the poet in the fictional source evidence. Completion of title migration and literal name matching do not establish correct human identity.’
The corrected process should preserve distinct creator identifiers under credible evidence, update affected public citations and retest same-name people alongside different-name variants of one person. The Gold Standard lesson is that finding every book is not enough when the catalogue sends readers to the wrong author.
Cataloguing authority references and further reading
- Library of Congress · Understanding MARC Authority Records
- LOC PCC · Authority Control FAQs for Authors and Creators
- LOC · Defining Source for Names and Titles in MARC 21
- PCC · Position Statement on Personal Data in Name Authority Records
- Library of Congress · Authorities Search
- The Gold Standard of Libraries · Discoverable Knowledge and Dependable Access
- Gold Standard · Information and Record Integrity
- Case 41 · Archive Migration Lost Retrieval Routes
- Case 113 · Museum AI Reconstruction Mislabelled as Photograph
- Gold Standard Casebook
Authority records and privacy choices require qualified cataloguing judgement. GS-R01–GS-R12 is eduKate’s educational management reference, not an external library accreditation.