[Push] Wondering about Code

[Rud Merriam]

I've been working on my own C++ implementation of Push for years. Started
with Push2 and am somewhere between it and Push3. Right now I just call my
work PushX - for unknown. Its been an off again / on again project but it
does get closer to completion. I'm actually getting to the point of trying
to implement code to do evolution.

I was just looking at the project again and re-read the Push3 documentation.

I am wondering why the CODE type still has the execution statements like DO,
DO*, etc. Keeping them for historical reasons is one possible good answer.
But it seems the main reason for CODE is as a workspace for manipulation of
code, rather than execution. EXEC is the place for execution in Push3.

I also wonder if the EXEC.Y, .S, and .K should be CODE instructions and not
EXEC. 

All these, I guess, are questions about regularizing or making the types
orthogonal. As a decades of experience developer I see this as good software
engineering. Shouldn't good evolved good follow good engineering practices?
(he says a bit tongue in cheek)

 
 - 73 - 
Rud Merriam K5RUD 
http://mysticlakesoftware.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.hampshire.edu/pipermail/push/attachments/20100414/2be9863b/attachment.htm>