[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: CLOS question



> Date: Thu, 21 Feb 91 12:44:40 EST
> From: cornell@unix1.cs.umass.edu (Matthew Cornell)
> .... 
> 
> On the m.o.p.: Is it standardized now? I thought the working group was
> still out on some issues. Where is it described? I could only find the
> first two chapters of the CLOS spec on ariza.xerox.com.

The metaobject protocol is not standardized.  Gregor Kiczales and Jim des
Rivieres, both of Xerox PARC, are publishing a book about metaobjects
this spring.  I'd tell you the publisher except I forgot where I wrote
it down, and don't want to give possibly false information by relying on
my memory.  This book consists of two parts, "The Art of the Metaobject
Protocol", which discusses concepts in a way of interest to a broader
community than just CLOS, and "The Metaobject Protocol Specification", which
is not intended to be a standard, but simply to document their thinking
so far and "form the basis for development of the improved CLOS Metaobject
Protocols of the future."  In other words, it has been recognized that
formal standardization of metaobjects would be premature at this time.
Fortunately, the work done so far is going to be made available in
an accessible form.  "The Art of" is quite well written (in the early
draft that I saw last fall) and much more readable than the past MOP
documents.  One needs to be careful about relying on this MOP book.  The
two sections do not describe the same language exactly ("The Art of" has
many simplifications for pedagogical purposes), and neither of them
describes the same language as anyone implements (although I think I
heard there was an effort to make a PCL that corresponds to the book.)

I'll let someone else more knowledgeable comment on the relationship of
this to Apple products.