Path: news.daimi.aau.dk!news.uni-c.dk!sunic!trane.uninett.no!eunet.no!nuug!EU.net!Germany.EU.net!news.dfn.de!swiss.ans.net!howland.reston.ans.net!math.ohio-state.edu!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!epflnews!monnier
From: monnier@di.epfl.ch (Stefan Monnier)
Newsgroups: comp.lang.beta
Subject: Re: unsafe virtual patterns?
Date: 22 Sep 1994 14:24:28 GMT
Organization: Ecole Polytechnique Federale de Lausanne
Lines: 19
Message-ID: <35s42s$pml@info.epfl.ch>
References: <35nlra$t2@rosie.next.com> <35nubg$74o@engnews2.eng.sun.com>
NNTP-Posting-Host: didec4.epfl.ch

In article <35nubg$74o@engnews2.eng.sun.com>,
Ole Lehrmann Madsen <olm@Eng.Sun.COM> wrote:
> it is not possible to get all 3 at the same time. For BETA we have
> decided to give up complete static typing. I.e. in the cases of
> covariance, the type-checking is deferred to run-time.

What's so great about covariance ?
On the programming side, I don't think it offers big improvments (it's
rather misleading to me, maybe because I think too much in terms of
type checking :-). On the implementation side, it forces run-time
checks or complicated (and moreover global) analysis.

In the Eiffel language, I accept the choice since they already have complex
multiple-inheritance and other stuff, but in a "simple and clean"
language like Beta, it amazes me.
Makes it harder to get C-like performance. 


	Stefan