The feasibility and the quality of such a catalogue strongly depends on the amount of work invested in rendering stellar identifications homogeneous. Stars can be called under a rather large variety of identifications, and many of them are in fact used. Therefore an important part of the action of cataloguing the data is devoted to the recovery of the best possible (primary) identification. The same problem is also encountered for stars in the region of star clusters and has been solved in an original way (Mermilliod 1979, 1986).
The existence of the database allows also to perform numerous checks on the data, by comparing the V magnitudes, plotting similar indices like (B-V, b-y) to search for anomalies and discrepant data. This may reveal interesting astrophysical questions, but more often uncovers a more or less trivial error. The origin may be difficult to trace back because it may happen at any stage of the chain, from the work at the telescope to the inclusion in the database. The first thing we check is that the data entered into the database are identical to those published.
Owing to its completeness, the GCPD offers wide possibilities for various applications. Usage may concern a statistical approach: (1) to locate the existing data for any star, (2) to prepare an observing campaign and identify the stars with or without data, (3) to investigate the completeness of available data for specific samples, like the nearby or bright stars, (3) to form a sample of stars measured in two or more photometric systems and study transformation relations, (4) to do any astrophysical research programme.
Our future plan is to implement existing calibration programmes on the database to allow users to get directly the astrophysical quantities through the Web server. Codes for the uvby and Geneva systems are available, one for the DDO photometry has been written and one for the Washington system has been received from J.J. Claria. We would be pleased to start any collaboration to implement additional software on the database and make the facilities available through the WWW server.
The main requirement of data archiving is to contribute to the improvement of the data quality and to the rejection of the poorest ones. It is the only way for disseminating reliable data and will provide future generations of astronomers with data that retain some interest. The effort made during long nights of observations by so many observers warrants some supplementary work to preserve the results and transmit them to posterity. Any correction or new information is welcomed to improve the database.
Acknowledgements
This long-term work has been supported by continuous grants from the Swiss National Foundation for Scientific Research (FNRS). We are pleased to thank Dr. Juan Claria (Cordoba, Argentina) for sending a copy of his code on the Washington system and Mrs. B. Wilhelm for reading and correcting the manuscript.