From: Mayka
Subject: Small object-oriented programming system for Lisp wanted
Date:
Message-ID: <8198@ihlpf.ATT.COM>
I run Austin Kyoto Common Lisp on my AT&T UNIX PC. I would like
to run an object-oriented programming system on top of AKCL.
Unfortunately, the UNIX PC architecture limits processes to
2.5 megabytes of virtual address space, so the Common Lisp
Object System will not fit. I am therefore looking for an
alternative object system with these characteristics (basically
in priority order):
a) Small enough to fit on the UNIX PC. AKCL comes up with
about 320 free pages, so the object system must fit within
about 600 Kbytes, compiled. Hopefully, the object system would
be *much* smaller than that, so that the AKCL compiler would have
room to compile Lisp programs that use the object system (at least
simple ones).
b) Freely available - e.g., via anonymous FTP or UUCP.
c) Using the style of generic functions rather than a specific
message dispatch function such as 'send' - i.e., the style
of CLOS and New Flavors, rather than Old Flavors.
d) Performing reasonably enough so as not to be totally dismissed by
naysayers as merely a toy.
e) Having some degree of use and popularity, at least in educational
settings. Documentation and examples would also then be more
likely to be readily available.
f) Having a syntax roughly similar to either CLOS or New Flavors.
Two candidates I know of, but would like to know more about, are:
a) Common Objects. I believe a Common Objects system, portable to
any Common Lisp, was once posted to the net. Where is it available?
Does anyone actually use it? How does it stand up to the criteria
above?
b) The object system used in the book, "LISP, Objects, and Symbolic
Programming," by Robert R. Kessler. The author not only uses it
throughout the book, he lists its source code in an appendix.
It looks quite serviceable, but I am loath to type it all in by
hand. Is it FTP-able from the University of Utah? I cannot find
the author's email address so I cannot easily contact him directly.
Any help in this matter would be much appreciated.
Lawrence G. Mayka
AT&T Bell Laboratories
···@ihlpf.att.com
From: Robert R. Kessler
Subject: Re: Small object-oriented programming system for Lisp wanted
Date:
Message-ID: <1563@wasatch.utah.edu>
In article <····@ihlpf.ATT.COM> ···@ihlpf.ATT.COM (Mayka) writes:
>
>I run Austin Kyoto Common Lisp on my AT&T UNIX PC. I would like
>.. [More text about the characteristics]
>b) The object system used in the book, "LISP, Objects, and Symbolic
>Programming," by Robert R. Kessler. The author not only uses it
>throughout the book, he lists its source code in an appendix.
>It looks quite serviceable, but I am loath to type it all in by
>hand. Is it FTP-able from the University of Utah? I cannot find
>the author's email address so I cannot easily contact him directly.
Ah, I should have put my email address in the Preface. Anyway, it is
·······@cs.utah.edu
>
>Any help in this matter would be much appreciated.
I would be happy to make the file publically available. Utah cs has
anonymous ftp available, all I need to do is copy the file out there
in whatever the appropriate directory is. I'll get do it as soon as I
can, so you and anyone else that wants a copy can pick one up. If
there are other parts of the book that you would like, I might be able
to extract them from the text (I LaTeX'd the entire book except for
the word LISP on the title page -- I couldn't get a font that big!!).
For example, I made the expert system avaiable to a colleague (who
promptly found a big which I haven't had time to track down -- any
takers out there?). SO, the files will be called bobjects.l (the b
prepended is because we had an objects.l from another system, so had
to give it something unique) and es.l for the expert systems chapter.
B.