Encoding Collective Knowledge, Instructing Data Reusers: The Collaborative Fixation of a Digital Scientific Data Set

Area of Science:

• Astronomy
• Data Science
• Sociology of Science

Background:

• Existing research on scientific data reuse often focuses on metadata.
• This study examines how data creators embed knowledge within the data itself for reuse.

Purpose of the Study:

• To provide an ethnographic account of preparing an astronomical catalogue for public release.
• To analyze how researchers designed the catalogue as an "instructing data object" to guide users.
• To explore the philosophical implications for collective knowledge and coordinative artifacts.

Main Methods:

• Chronological ethnographic analysis of a research team's data preparation process.
• Examination of methods used to structure anticipated data uses, including redundancies and selective deletion.
• Analysis of notational choices for accountability and knowledge encoding.

Main Results:

• The catalogue's final form was a negotiation among team members, balancing usability and scientific coherence.
• Researchers embedded processing knowledge and anticipated uses directly into the data object.
• Methods like redundancy and selective deletion were employed to guide users and manage accountability.

Conclusions:

• Scientific data can be prepared as "instructing data objects" to facilitate reuse and encode collective knowledge.
• This approach highlights a tension between algorithmic processing and scientific practice in data preparation.
• The study offers insights into collective knowledge and coordinative artifacts in Computer-Supported Cooperative Work (CSCW).