Besides programming effort, is there a reason why squeak/pharo shouldn't consider going to a GUI model that includes real multiple native windows? I'm not a c/c++ programmer and wouldn't know where to begin with that piece, but inside of smalltalk it seems it wouldn't be all that difficult of a refactoring. I only bring this up as this would put pharo/squeak more in line with many other conventional languages (with respect to user interfaces), and get it "back in the game" so to speak for when technologies are being considered for business solutions. You could include in this 1) Native support for drag and drop and 2) support for embedded native controls. Something like SWT and Jide now both support. Sorry if I've overlooked something with this and the con is obvious. thanks sw