Re: Functional OO

Graham Matthews <graham.matthews@wintermute.anu.edu.au>
30 Apr 1996 23:58:07 -0400

          From comp.compilers
Related articles
Functional OO mtimmerm@microstar.com (1996-04-29)
Re: Functional OO hamel@Think.COM (Lutz Hamel) (1996-04-30)
Re: Functional OO graham.matthews@wintermute.anu.edu.au (Graham Matthews) (1996-04-30)
Re: Functional OO ranjit@fwasted.zk3.dec.com (1996-05-01)
Re: Functional OO patrick_d_logan@ccm.hf.intel.com (Patrick Logan) (1996-05-01)
Re: Functional OO iainf@bristol.st.com (1996-05-06)
Re: Functional OO hans@iesd.auc.dk (Hans Huttel) (1996-05-10)
| List of all articles for this month |
From: Graham Matthews <graham.matthews@wintermute.anu.edu.au>
Newsgroups: comp.theory,comp.compilers
Date: 30 Apr 1996 23:58:07 -0400
Organization: Australian National University
Distribution: inet
References: 96-04-157
Keywords: OOP, functional

Matt Timmermans wrote:
> Is there any known notion of functional object oriented semantics? Is
> there any side-effect free language that could be called object
> oriented?
>
> I know there are difficulties with such an idea, but are these two
> paradigms inherently incompatible?


Why should there be any difficulties fusing OO and functional
languages. You won't be able to fuse the imperative part of OO,
namely arbitrary obect updates, into a functional system, but that
doesn't mean you can't have the rest of the OO feature
set. Inheritance, encapsulation, etc are all independent of whether
you have a functional or imperative semantics. Moreover things like
memoisation get you some way toward "object updating".


graham
--

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.