Blog

Articles on automated cataloging, catalog formats and use cases.

Authority control: each name enters the catalog in one form

The same author spelled three ways is three people in the index. Collect checks each name and subject against the lists the institution already uses —its catalog, VIAF, ISNI, the Getty vocabularies— and records which list the form came from.

Cataloging a collection video with AI: from the recording to linked records

A newsreel or a recorded session does not come with the content written down. Collect proposes a record for the whole video and, if needed, a record per scene, linked to the first.

Access points, full text and RAG

A recent study measures how much context a model actually uses. What that means for RAG, and what an institution gets when Collect surfaces access points —including the corporate extensions— for Janium to search.

Collect marks AI-made metadata

Collect meets Article 50.2 of the EU AI Act: cataloguing output carries a machine-readable mark, the catalog record does not change, and the proof stays with the institution.

From a sound recording to a catalog record

How Janium Collect catalogs collections with audio —sound recordings, interviews, oral minutes— transcribing the voice and producing a record in Dublin Core, MARC or ISAD-G, with the option to segment by intervention into linked child records.

From loose documents to Fonds and Series: generating the ISAD-G hierarchy automatically

How to automatically generate the ISAD-G hierarchy —Fonds, Section, Series— while cataloging: the exporter builds the tree from the classification scheme, raises to the upper levels what most of the children share, and adds date range and extent.

What to demand from an AI cataloging tool

Almost any AI cataloging tool looks good in a demo. These are the questions that reveal whether its records are reliable in your catalog, beyond that first impression: where each datum comes from, against which authorities it verifies, what it computes by rule, whether it is correct against the standard, whether it evaluates each record, whether it fits your collection, whether it respects your sovereignty and whether it is honest about its limits. They work for evaluating any vendor.

Colegio de La Salle: from a four-field Excel to a library system in the cloud

The library of a bilingual school in Bogotá kept its collection in a spreadsheet with four columns per book. How that list became MARC21 records with Collect, and a system that today catalogs, circulates and publishes its catalog online.

The Diocesan Library of Bilbao: cataloging a vinyl collection from a photograph of the record

An old collection of sound recordings —vinyl records of religious music in Spanish, Basque and Latin— is hard to catalog because its information lives on the sleeve, not in a database. How the library cataloged it with JaniumCollect from photographs.

Generic isn't enough: why a catalog of record needs more than an LLM that fills fields

A generic LLM already produces complete, plausible records; enriching is the value of AI in cataloging. What a catalog of record needs on top of that is knowing where each datum comes from: the source that backs each one, deterministic computation where there's a rule, per-record evaluation and flagging, the logic of the standard, multi-language. Along with the limits of enrichment from the model's knowledge.