Last Call Issue: WORKUNIT REPETITION

Suggestion from Kohei based on his analysis of the workunit semantics for repetition:

> 
> We therefore believe it is better to have two workunits, one for non-repeating,
> and another for repeating. If this is too radical, then we recommend not to have
> two conditions, but to differentiate between repeating and non-repeating by:
> 
> repeat = true|false 

Regards
Gary and Steve

Received on Wednesday, 5 January 2005 14:19:07 UTC