home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!mucs!cs.man.ac.uk!mod
- From: mod@cs.man.ac.uk (Mike O'Docherty (Teaching Company Associate with MU/ICL))
- Newsgroups: comp.lang.eiffel
- Subject: Re: system-level validity
- Message-ID: <mod.715877622@cs.man.ac.uk>
- Date: 7 Sep 92 14:53:42 GMT
- References: <mod.714411318@cs.man.ac.uk> <714585601snx@holon.demon.co.uk> <mod.714907485@cs.man.ac.uk> <83@feiffel.UUCP> <1bmn50.nagle@netcom.com>
- Sender: news@cs.man.ac.uk
- Lines: 20
-
- nagle@netcom.com (John Nagle) writes:
-
- > Given system-level validity checking, what strong statements can
- >one make about a program that passes it?
-
- I don't know if bland statements such as
-
- "A system-valid application is *guaranteed* to be type-safe, a
- class-valid system is not."
-
- are any use to you. No? Oh, well.
-
- Chapter 22 in ETL explains the problems in full. If you can't get hold of a
- copy then you'll just have to appreciate that it is *possible* to
- access a feature on an object which either doesn't have that
- feature or doesn't export it to you or expects different arguments.
-
- It's a complex business this type-checking!
-
- Mike.
-