fixed DESIGN-NOTES conflict

svn: r1320

notes/stepper/DESIGN-NOTES

@@ -803,6 +803,28 @@ outside in?  Feh.
 2005-08-02
 
 
+*************
+
+Well, the dang summer is almost over, and I've still got a long, long way to
+go.
+
+The basic change to the model is that instead of storing completed definitions
+as pre-formatted s-expressions, I'm now storing them as 2-element lists
+containing the syntax object associated with the definition and a 'getter'
+which returns the value that the binding refers to.  The actual definition is
+reformatted for each step.  This is a bit silly, but it would be easy to cache
+the definitions along with the present values if this is actually a performance
+bottleneck. I suspect it won't matter a bit.
+
+In the presence of mutation, the existing separators don't make sense, either.
+I'm scrapping them, for the moment.  A nice interface change would be to
+separate only the definitions that had changed. For them moment, they'll all be
+separated.
+
+The first order of business, after mucking around in the model for some time to
+get the flavor of how things will work, is to go and set up the interface so I
+can get things running.
+
 *************
 
 Okay, I've "completed" the google project, but there are still things
@@ -841,4 +863,9 @@ hide in the "syntax?" contract.
 
 **************
 
+Time to clean up for v300.  Let's see if we can get begin and begin0 working.
+
+2005-11-14
+
+*************
 

notes/stepper/OPENBUGS

@@ -1,3 +1,7 @@
+The reduction of 'recur' does not conform to a sensible reduction
+semantics.  Can you figure out why?  (For the answer, see the comment
+in the 'recur' section of collects/tests/stepper/through-tests.ss.)
+
 Primitives like 'make-number/make-numeric' that are renamed
 as module provides are not displayed using their renamed
 names.