For those following the debate on the implementation of the Rete algorithm in the Microsoft Business Rules Engine, you may be aware that I undertook to port some CLIPS/Jess tests provided by Peter Lin to the MS-BRE and to publish the results. This follows on from the original article I published here.
My results for the ported tests are documented here. They show virtually identical scaling behaviour between Jess and the MS-BRE.
There has been further debate about this on The Server Side .NET site, which can be read here.