Newsgroups: comp.lang.java,comp.object,comp.object.logic
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!newsfeed.internetmci.com!howland.reston.ans.net!psinntp!psinntp!psinntp!psinntp!twratl!tynor
From: tynor@atlanta.twr.com (Steve Tynor)
Subject: Re: Java public instance variables harmful?
In-Reply-To: bobduff@world.std.com's message of Mon, 18 Mar 1996 22:35:25 GMT
Message-ID: <TYNOR.96Mar18210548@twratl.atlanta.twr.com>
Sender: tynor@atlanta.twr.com (Steve Tynor)
Organization: Tower Technology
References: <KRUSE.96Mar6131047@cms6.cern.ch> <4hkhnv$5ne@decaxp.harvard.edu>
	<4ht4a9$35@gaia.ns.utk.edu> <DoHIr2.A78@world.std.com>
Date: Tue, 19 Mar 1996 02:05:48 GMT
Lines: 31
Xref: glinda.oz.cs.cmu.edu comp.lang.java:32510 comp.object:46187 comp.object.logic:758

In article <DoHIr2.A78@world.std.com> bobduff@world.std.com (Robert A Duff) writes:

| In article <4ht4a9$35@gaia.ns.utk.edu>,
| RHS Linux User <kennel@msr.epm.ornl.gov> wrote:
...
| >Better yet, use a language which doesn't have this quite
| >this silly distinction between 'variables' and 'accessor methods'.
| 
| I agree.  We need to figure out how to avoid efficiency costs, though.

There's no efficiency cost from redefinining a routine as a variable
in Eiffel. In our implementation, it's a net efficiency _gain_ in
those contexts where the compiler knows that the feature is
implemented as a variable.

| >Like Eiffel, which lets descendents re-implement methods as variables
| 
| OK.
| 
| >and vice versa. 
| 
| Last time I looked, that was not allowed in Eiffel.  (The vice versa;
| that is, the implementation of variables as methods.)  True?

True.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Programming needn't be like penance or oat bran.

Steve Tynor		Internet: Steve.Tynor@atlanta.twr.com
Tower Technology 	WWW:      http://www.twr.com/
