Thank you for your project submission. I give you some feedback below. Please can you respond and update your submission as you see fit. I would like to have from each of you a message which gives your estimate of work distribution, for example: Jensen 40% Wagstaff 30% Danielson 30% and maybe some oral explanation. It is great that you helped each other especially since one of you had such a challenging personal life. Your project is _very_ interesting. You use a schema as a network directory. This opens a new application area for adaptive programming. You probably experience frequent changes in the schemata and a schema-shy approach ought to simplify the NDS task. Maybe we should have a discussion on this? I assume it is ok to post your project on the web provided all your copyright notices and disclaimers are there? Please shorten your cd. Write only ONE List(S) ~ {S}. // this is like a template class in C++ and then use it like ... List(SyntaxDef) ... the way you already do. Get rid of: List(SyntaxDef) ~ { SyntaxDef }. List(SyntaxFlags) ~ { SyntaxFlags }. ... Is the textual input interface the same for the Demeter/Java and C++ version. Or do you have a special input language for the Java version? What is your motivation for: SuperClasses = String. ? I would expect: SuperClasses = List(ClassName). A similar pattern happens several times. What is your motivation for: traversal toClassDef(ClistVisitor cV) {bypassing {AttrList,SyntaxList} to ClassDef; } when both AttrList and SyntaxList cannot contain a ClassDef-object? The bypassing seems not to be necessary. The same pattern of unnecessary bypassings exists several times. In your cd, it is better to use List(ClassDef) instaed of ClassDef_List. Use your parameterized List class systematically to avoid maintenance problems. Same pattern repeats several times. I checked your white paper but I found nowhere an explanation how your schemas use syntax. Is syntax used in a similar way as in Demeter class dictionaries? I tested your WWW links http://www.ccs.neu.edu/research/demeter/course/f97/projects/novell/project but many of them don't work. Unix is case sensitive. Please can you correct them. How does your program compare to another schema comparer in: http://www.ccs.neu.edu/research/demeter/course/f97/projects/object-equiv I tried to run the C++ executable on my PC but it was missing a file. How much slower is the Java version than the C++ version?