Xref: utzoo comp.object:3436 comp.lang.misc:7698 comp.lang.eiffel:1565
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wuarchive!cs.utexas.edu!asuvax!noao!arizona!optima.UUCP
From: cjeffery@optima.UUCP (Clinton Jeffery)
Newsgroups: comp.object,comp.lang.misc,comp.lang.eiffel
Subject: Re: A Hard Problem for Static Type Systems
Message-ID: <2672@optima.cs.arizona.edu>
Date: 2 May 91 09:53:35 GMT
References: <566@eiffel.UUCP>
Sender: cjeffery@cs.arizona.edu
Followup-To: comp.object
Lines: 17

From article <566@eiffel.UUCP>, by bertrand@eiffel.UUCP (Bertrand Meyer):
> If we were able to build static checkers that were totally
> unobtrusive performance-wise, and did their work in - say -
> ten seconds after a comparatively small change even to a very large
> system, then who in the world would forsake the extra benefits of type
> checking?

I would.  I am not willing to type one keystroke (e.g. type declarations)
more than I have to in order to satisfy your need for everyone to do so.
What is this sweeping generalization doing here after your very nice
concession to dynamically typed languages earlier in your post?

The closest I am willing to come to your cumbersome world is to let my
type inference system insert comments into my code when it notices I am
using a variable for different types of values at different times.

Keystrokes.  How many keystrokes does it take me to solve my problem?