Russ Abbott wrote:
I think I finally put my finger on how to characterize what's confusing about Oz. Let me know what you think of this: http://cs.calstatela.edu/~wiki/index.php/Courses/CS_460/Fall_2005/What_programmers_should_know_about_Oz <http://cs.calstatela.edu/%7Ewiki/index.php/Courses/CS_460/Fall_2005/What_programmers_should_know_about_Oz>. It is the second part, what I call the Oz communitarian run-time environment, that is particularly important. CTM does make it clear from the start (1.11) that Oz operates this way. What I don't think it does is to clarify how different this makes programming in Oz from programming in other languages. In 1.11, dataflow variable seem like a cute trick. They are far more significant than that, and they have far stronger implications about how to understand the semantics of Oz programs. That should be brought out and stressed, and programmers should be warned about how often and in how wide a range of circumstances threads suspend. -- Russ
In CTM see section 2.2.8 Dataflow Variables (page 47) for an explanation of how dataflow variables make programming different from other languages. See also page 89 'Dataflow execution' that makes exactly the point that you are making. Peter _________________________________________________________________________________ mozart-users mailing list [email protected] http://www.mozart-oz.org/mailman/listinfo/mozart-users
