Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site watcgl.UUCP
Path: utzoo!watmath!water!watcgl!jchapman
From: jchapman@watcgl.UUCP (john chapman)
Newsgroups: net.micro.pc,net.legal
Subject: Re: Boiler plate warranties - software isn't that reliable
Message-ID: <2491@watcgl.UUCP>
Date: Tue, 10-Sep-85 10:02:08 EDT
Article-I.D.: watcgl.2491
Posted: Tue Sep 10 10:02:08 1985
Date-Received: Wed, 11-Sep-85 05:12:52 EDT
References: <3199@pur-ee.UUCP> <332@looking.UUCP> <1082@mtgzz.UUCP>
Organization: U of Waterloo, Ontario
Lines: 62
Xref: watmath net.micro.pc:5257 net.legal:2301

> > In article <3199@pur-ee.UUCP> kk9w@pur-ee.UUCP (David Andersen) writes:
> > >
> > >The fact that someone wrote that they are not responsible does
> > >not mean that they are not responsible.  If something breaks
> > >that a reasonable person thinks shouldnt, regardless of the
> > >boiler plate on the wrapper the manufacturer is liable for
> > >damages.  If my software screws up and causes me a substantial
> > >loss, the software manufacturer could be liable for those
> > >damages, possibly to the tune of 3M$ or more. 
> > 
> > 
> > You have got to be kidding here.   Software just isn't that reliable.
> > Whatever warranty system you use, you have to take into account how it
> > will be in practice.  All programs have bugs, and only massive redundancy
> > and great efforts can ensure software as safe.  Big companies might be
> > able to afford it, but you will kill the little guy requiring liability
> > like this.

   If so much of the software from small outfits wasn't complete trash
   you might have a point; disclaimers that the manufacturer doesn't
   warranty the program to do anything are bs as far as I as a consumer
   am concerned.

> > 
> > Think about doctors and malpractice insurance.  Think about car insurance.
> > Immense awards are driving up the price to the consumer.  Some doctors
> > won't deliver babies anymore because of the price.
> > 
> > Some liability and warranty may be in order, but this magnitude is too
> > much if you want the little guy to be able to produce software.  Customers
> > who buy software must realize how complex a task writing it can be.
> > If I sell a C compiler, I could never possibly imagine all the different
> > kinds of code that might be written with it.  There might be safe things

    These sorts of problems are one very good reason for the use of highly
    structured, strongly type checked languages like Pascal & Modula-II.
    The use of a well designed validation suite is also a good idea.
    You do not have to imagine all the possible applications programs
    to verify all the possible output constructs a well written and
    structured compiler will produce.  There is also a question of degree:
    in ten years of using IBM mainframe software (I'm not particularily an
    IBM fan btw) I suffered because of a software bug.  This has hardly been
    the case with micro software.

> > or there might be a traffic control system.  If somebody writes a traffic
> > control system that causes a car crash because of a bug in my compiler,
> > am I liable?  Or do I have to put, "this compiler is suitable for compiling
> > only safe applications software that doesn't control physical machinery"
> > in my advertising?

    National Semiconductor puts just such a warning on some of their products
    ("not to be used for life support applications").
> > 
> > -- 
> > Brad Templeton, Looking Glass Software Ltd. - Waterloo, Ontario 519/884-7473
-- 

	John Chapman
	...!watmath!watcgl!jchapman

	Disclaimer : These are not the opinions of anyone but me
		     and they may not even be mine.