A New Approach to Tagging Data in the Astronomical Literature

Abstract

Data Tags are strings used in journals to indicate the origin of the archival data and to enable the reader to recover the data. The NASA/IPAC Infrared Science Archive (IRSA) has recently introduced a new approach to production of data tags and recovery of data from them. Many of the data access services at the IRSA return filtered data sets (such as subsets of source catalogs) and dynamically created products (such as image cutouts); these dynamically created products are not saved permanently at the archive. Rather than tag the data sets from which the query result sets are drawn, the archive tags the query that generates the results. A single tag can, then, encode a complex dynamic data set and simplifies the embedding of tags in manuscripts and journals. By logging user queries and all the parameters for those query as Data Tags, IRSA can re-create the query and rerun the IRSA service using the same search parameters used when the Data Tag was created. At the same time, the logs give a simple count of the actual numbers of queries made to the archive, a powerful metric of archive usage unobtainable from the Apache web server logs. Currently, IRSA creates tags for queries to more than 20 data sets, including the Infrared Astronomical Satellite (IRAS), Cosmic Evolution Survey (COSMOS) and Spitzer Space Telescope Legacy Data Sets. These tags are returned by the spatial query engine, Atlas. IRSA plans to create tags for queries to the rest of its services in late Spring 2007. The archive provides a simple web interface which recovers a data set that corresponds to the input data tag. Archived data sets may evolve in time due to improved calibrations or augmentations to the data set. IRSA's query based approach guarantees that users always receive the best available data sets.

Files

Additional details