On Tue, 24 Mar 1998, Thierry Goubier wrote:
Long transactions, nested transactions, a transaction-based squeak up to the UI (morphic or MVC). (of course, the transactions boundaries are automatically managed by the system to increase the parallelism between processes and with others VMs. Distributed transactions, too)
Just dreaming... But I'd like to be proven wrong...
So, any thoughts on how to exploit self similarity in a persistence model? I'm not really sure where to take this idea. Would it be possible to improve performance if more kinds of objects could be mapped onto fewer database tables?
Carl ------------------------------------------------------------------ Carl Gundel carlg@world.std.com Shoptalk Systems 508-872-5315 author of Liberty BASIC, a 1996 PC Magazine Awards Finalist! http://world.std.com/~carlg/basic.html ------------------------------------------------------------------