Acknowledgments

Part of this work has been supported by the EU-funded (FP6) Euro-VO VOTECH project.