Originally created in 1980 as the Guide Star Selection System (GSSS), the primary goal was to provide positions of guide stars to point the Hubble SpaceTelescope (HST).

The GSC-I was constructed primarily to support the operational requirement of the Hubble Space Telescope for off-axis guide stars. HST achieves its pointing and stabilization by the use of pairs of guide stars in the range 9.0 - 14.5 mag. The criteria for selection of guide stars are set, in part, by the small field of view of HST's Fine Guidance Sensors, by the sizes of science instrument apertures, and by the precision of the inertial system provided by the gyroscopes.

The initial strategy (c. 1979) was to create a production astrometry system driven by the HST observing schedule. Two microdensitometers would be used to scan the specific fields of photographic plates for which guide stars were needed, and subsequent reductions would provide a list of possible guide stars. These plans would have required a three shift per day operation over the operational life of HST, resulting in marginally adequate throughput for normal observing and slowing response time to targets of opportunity.

To overcome the problems associated with this initial strategy, we adopted the approach of scanning all the plates once at the beginning of the program, archiving the digital images, building a star catalog, selecting possible guide stars from the catalog, and developing utilities to solve operational problems such as identification of faint targets. To save time, this plan was based on scanning the photographic plates with 50um (3.4") samples. This would have required additional fine scans of guide star candidates and targets at 25um sampling to determine more accurate positions for observations requiring better astrometry.

Optimization of the microdensitometers for full plate scans, combined with a delayed HST launch, led to the final plan to digitize all the plates at 25um sampling, eliminating the need for partial rescans of many fields and leading to the construction of GSC-I with its ultimate form and precision. Early design studies also featured measuring at least two plates per field to provide proper motions and colors. The final design minimized proper motion errors by acquiring recent-epoch plates and minimized magnitude errors by using passbands closer to that of the HST Fine Guidance Sensors.

Production scanning began in 1985 and continued through 1988. Catalog construction was performed in parallel with the scanning. It proceeded in stages as production software became available, first for sky-following and object inventory, and later for classification plus photometric and astrometric calibrations.

Once the GSC-I was in place and successfully pointing HST, our group realized the operational and scientific utility of access to the all-sky images. We developed a compression algorithm which reduced the image data volume by ~10x and published the "Digitized Sky Survey" for community use on 102 CDROMs.

As operational needs evolved, the next logical step was to create a "bigger, better" catalog! The GSC-II