I have started working on a request for new facilities for RAS. As I am not familiar with cloud computing, and Thomas suggested we use that, I have read some about it. I do not quite see why we would want to use that. My understanding is that its main benefit is to avoid buying and maintaining hardware. This is useful when 1) you do not have the means for big expenses, 2) you do not have the ability to maintain systems and 3) if your computing power needs are very variable over time. You can then avoid spending much for peak loads. I have little worry for 1) and 2), as the St. Louis Fed can buy and maintain. As for 3), our needs are much less cyclical than we think, and by doing it intelligently, can be completely smoothed by making automatic searches load dependent. Cloud computing may make more sense for the centrality project, but given that it uses the same data as RAS, it would actually make better sense to put it on the same machine as well. Hence, I am tending towards requesting a really powerful machine that could handle RAS and centrality, trying to figure out the specs needed for 2 million templates, 50000 authors and 1.2 million author claims, essentially double what we have now. I akm thinking of adding some essential new functionalities to RAS, but none would require any substantial coumputer resources. Comments? Christian Zimmermann FIGUGEGL! Department of Economics University of Connecticut 341 Mansfield Road, Unit 1063 Storrs, CT 06269-1063 http://ideas.repec.org/zimm/ christian.zimmermann@uconn.edu http://ideas.repec.org/e/pzi1.html