|SPNHC 2014: Progress in Digitization: Designing a Workflow to Help with Error Detection in a Paleontology Digitization Project
|Year of Publication
|Mayer, Paul S., Connolly L., Karpus N., and Layng A. P.
|digitization, error checking, Field Museum, fossils, interns, Silurian, SPNHC 2014, workflow
|The Field Museum is implementing a three-year IMLS grant using summer interns to digitize approximately 15,000 Silurian fossil lots. Workflow for this project has each intern working on six drawers of fossils at a time; the fossils are all from the same geologic period and taxonomic group (as they are arranged in the collection). Interns cycle their fossils through three workstations: a label photography station, a fossil photography station, and a KE EMu data entry station. The average time per specimen to complete this cycle is 9.5 minutes with times ranging between 4 to 18 minutes. After cycling their cart through each workstation I edit, upload, and link the images to the catalogue records. In the first year 7,785 image were generated and linked to catalogue records.
During the process of connecting images to the catalogue there are three opportunities to check for errors. First, KE EMu generates an error log of images that fail to link during the batch input. The error rate is less than 2.2%. Common errors include simple format errors in the batch upload, specimens not entered into the catalogue, and typos in the specimen’s catalogue number. Second, because specimens in each batch are from the same geologic period and taxonomic group it is easy to spot any inconsistencies or missing data when reviewing the records in a table format. Finally each record should have two images linked to it. Any missing images are easy to add.