Back To Schedule
Thursday, May 31 • 5:00pm - 5:30pm
(Photographic Materials) How to Receive and Organize a Collection of 1 Million Photographs at Once? Material and Metadata Discussions

Log in to save this to your schedule, view media, leave feedback and see who's attending!

All over the world for decades, newspapers and journalist groups have formed huge collections of photographs and clippings. This presentation aims to analyze the strategies used and projected to describe the photographic collection of the Jornal do Commercio (Comercy newspaper). When this newspaper was closed in 2016 it has been the longest paper in activity in South America; in the same year, Instituto Moreira Salles, a cultural institute that stores documental collections, bought the collection of photographs that was gathered by this newspaper. Over 70 years the newspaper collected around 1 million photographs, 700,000 photos, and 300,000 negatives, most of them about Brazil. These pictures were stored in file folders that received a thematic title in order to organize the photos in series. Much information about the pictures was registered on the back part; it’s possible to identify information such as date, place, photographer, newspaper where it was published and sometimes the full article. In order to catalog this collection, from the hugest series to each photograph, it was necessary to identify ways to transcript all the available data. The 1 million photographs were kept in around 1900 boxes; each of them contained 2 to 120 cardboard files; in the top of these files there is a title that informs which kind of pictures are in the files. In the original organization the collection was divided in two huge series, subjects and personalities; those series were divided into thousands of smaller series. The first tool used was a penscanner that can scan digitized texts and apply OCR, but this scanner wasn’t useful and precise in the old cardboard files, because there was no contrast. This pen only has good results in white paper. Then the team started to use a voice recognition software available in any Macintosh operational system. This software, used in Portuguese, reached high levels of precision and helped to make the process of description of the series very quickly. This software couldn’t be used in the personality series, because it only works in one kind of idiom, so the team is taking pictures of the cardboard files in order to apply OCR. The digital capture of useful informations for cataloguing and to describe this collection is a strategy to register the documents of the collection in a fast and accurate way. It’s also a conservation initiative, because it avoids information and documents disassociation. This set of actions have an important role to insert the cataloguing data in international standard like Dublin Core, Lido and ISAD(G). In addition discussions related to thesaurus, folksonomy and automatic indexing are equally relevant for this works and strategical group of actions. 

avatar for Rodrigo Bozzetti

Rodrigo Bozzetti

Registrar / Historian, Instituto Moreira Salles
Graduated in Library Science in 2012, by the Federal University of Rio de Janeiro state. Master degree in Information Science in 2016 by the Brazilian Institute of Information in science and Technology, where I developed an epistemological study about the concept of document. Since... Read More →

Thursday May 31, 2018 5:00pm - 5:30pm MDT
Hunters Creek Meeting Room Marriott Marquis Houston