> > If the server returns the old resultset name then the set has been > > rewritten. > But do we want to allow result sets to be re-written? You could argue that > violates the result set model. Of course real Z39.50 allows it, but gets > around the model issue by saying that the old result set is deleted and a new > one by the same name is created. Maybe it's simpler just to say "create a new > result set". Leave unanswered the question of whether the old one stays > around. Can't we use the same 'model' in SRW to allow it? Then we can have all three possibilities. (And just map z39.50 resultsets to SRW resultsets) For other messages: The combining resultset method is good :) And Ray has it right about what I meant for failed sorts. Rob -- ,'/:. Rob Sanderson ([log in to unmask]) ,'-/::::. http://www.o-r-g.org/~azaroth/ ,'--/::(@)::. Special Collections and Archives, extension 3142 ,'---/::::::::::. Twin Cathedrals: telnet: liverpool.o-r-g.org 7777 ____/:::::::::::::. WWW: http://liverpool.o-r-g.org:8000/ I L L U M I N A T I