About explicit metaclasses and old experiences with them

Stephane Ducasse ducasse at iam.unibe.ch
Mon Sep 14 13:33:54 UTC 1998


Hi squeakers,

I'm reading the History of Programming Languages paper on 
the birth of Smalltalk by Alan Kay. Really interesting. Read it....

I have some questions for the former members of the Xerox Team.

Some context: my background "was" CLOS & MOP inside with a clean
metaclass model with explicit metaclasses. So every class could be an
instance of the same metaclass and you do not need to reimplement all
the time a singleton metaclass for example. I also implemented several
times this tiny but beautiful ObjVlisp model.

An important remark is that these beautiful models do not support
explicit metaclass composition. So yes they are simple and suited for education 
especially ObjVlisp but they do not deal with the real issue: composition. 

I was writing slides for a presentation on the meta architecture proposed 
in Smalltalk and I want to know:

1- if the Xerox team tried some explicit metaclass approaches?
   (I'm finishing to read the HOPL Kay paper)
2- Is there a report, paper.. that discussed the problems?
   Why the current architecture was chosen? 
3- Having to explain to my audience the smalltalk meta architecture
   I find it a bit difficult for children understanding (ObjVlisp is already 
   difficult for master student ;)
   NeoClasstalk is the only language that I know that is proposing 
   a clean composition for explicit metaclass (see next OOPSLA proceedings) 
   
   What is the plan of the Squeak team to deal with that aspect of the 
   language? 


Stef


Stephane DUCASSE (ducasse at iam.unibe.ch) http://www.iam.unibe.ch/~ducasse/
"if you knew today was your last day on earth, what would you do 
different? ... especially if, by doing something different, today 
might not be your last day on earth" Calvin&Hobbes





More information about the Squeak-dev mailing list