Re: parallelism

[Marcus G. Daniels]

>>>>> "RS" == Ralf Stephan <address@hidden> writes:

RS> I see the benefit of the message syntax but this would require
RS> radical steps away from the C heritage for people accustomed to it
RS> through Java/Obj-C but hey, it's time.  I think it would help to
RS> have an accelerated version of the [myActivity execute...] or the
RS> setter methods like #pseudodefine XYZ:=A self.setXYZ(A) but I can
RS> offer no compiler experience nor any educated opinion on
RS> maintenance cost here.

I'm beginning to accept Glen's original implication that it is okay if
in some cases the modeler has to take care to keep probing machinery
up-to-date.  I don't accept it in the context of straight coding like
we do with Swarm now¹, but in framework for building models, we can create
a gradient toward small code fragments where side effects occur through
a common, well-understood interface.  I should have a proposal up in
a few days about such a framework.

¹ debugging is hard enough with C!  If we kept doing things the way
we do now, I would want to see a OS-level or language `write barrier'.






                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================