Newsgroups: comp.lang.c++,comp.object,comp.theory,comp.lang.smalltalk
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!cam-news-feed3.bbnplanet.com!cam-news-hub1.bbnplanet.com!news.bbnplanet.com!news.maxwell.syr.edu!news.bc.net!unixg.ubc.ca!van-bc!n1van.istar!van.istar!west.istar!news-w.ans.net!newsfeeds.ans.net!news.phx.mcd.mot.com!schbbs!news
From: shang@corp.mot.com (David L. Shang)
Subject: Re: Opinions on Ellipse-Circle dilemma?
Reply-To: shang@corp.mot.com
Organization: MOTOROLA 
Date: Fri, 14 Feb 1997 16:06:32 GMT
Message-ID: <1997Feb14.160632.10016@schbbs.mot.com>
References: <3303611e.0@news3.paonline.com>
Sender: news@schbbs.mot.com (SCHBBS News Account)
Nntp-Posting-Host: 129.188.128.126
Lines: 47
Xref: glinda.oz.cs.cmu.edu comp.lang.c++:247918 comp.object:61158 comp.theory:17965 comp.lang.smalltalk:51620

In article <3303611e.0@news3.paonline.com> mfinney@lynchburg.net writes:
> In <5dtcui$bik@bagan.srce.hr>, kmajor@jagor.srce.hr (Kazimir Majorinc)  
writes:
> 
> >: For details in theorectic aspects, language descriptions, and
> >: more examples, the following papers are available:
> >
> >:  1. Covariant Deep Subtype Reconsidered
> >:     ACM Sigplan Notices, May, 1995.
> 
> There are many more available than that, and the opinion expressed 
> in that paper about covariant parameters is far from being undisputed.

Please be specific.

> And as far as the assertion that covariant specifications are unsound
> there has been a good deal of work showing how the formalisms can
> be made sound.  In particular, the book...
> 
>    A Theory of Objects
>       Martin Abadi & Luca Cardelli
>       0-387-94775-2
> 
> which was recently published, considers theories for both first and
> second order constraints (such as covariance), where the theories
> are sound for the second order constraints.
> 

The second order constraints is basically equivalent to the contraints
on class parameters. The solution presented in
"Covariant Deep Subtype Reconsidered" (published two years ago) exactly
uses covariant contraints on class parameters, rather than on object
parameters (first-order constraints).

Further, the second order constraints presented in "A Theory of Objects"
does not deal with dynamic binding, a run-time polymophism that we
usually should have for the first-order constraints. If dynamic binding
was disallowed at all for the first-order constraints, the first-order
constraints would be sound too.

The language Transframe (http://www.transframe.com) provides a dynamic
parametrism, which is not the dynamic parametrism as in the form
"for all t<:T t->t" in other thoery works, allowing a dynamic binding
of class parameters to create new subtypes at run-time, and hence
make the constrains on the second-order dynamic.

David Shang
