Subject: Re: EROS and liability
Date: Wed, 30 Jun 1999 11:13:59 -0400

>I guess I just don't find this case convincing.  I think the moment
>the manufacturer uses different source code, the supplier of the
>original source code is clearly no longer liable.

What matters is what a jury/judge thinks.

The empirical evidence is that a judge/jury *will* accept what Ian says under
two conditions:

1. There was the clearly stated contract.  I.e. I sold a contract that said "you
modify it and you take all the burden"

2. I can prove that you modified it.

The sad reality is the burden of proof in this situation will fall on the
defendant, thus my question about methods for determining if modification
occurred in the face of configurable binaries.

Jonathan S. Shapiro, Ph. D.
IBM T.J. Watson Research Center
Phone: +1 914 784 7085  (Tieline: 863)
Fax: +1 914 784 7595

Ian Lance Taylor <> on 06/29/99 10:30:44 PM

cc:    (bcc: Jonathan S Shapiro/Watson/IBM)
Subject:  Re: EROS and liability

   From: "Stephen J. Turnbull" <>
   Date: Wed, 30 Jun 1999 11:13:05 +0900 (JST)

   >>>>> "Frank" == Frank Hecker <> writes:

       >> Secondly, I still just don't see any particular free software
       >> aspect to this.  I think vendors of proprietary RTOS's
       >> distributed in source code form have exactly the same problem.

       Frank> I'd agree.  Actually, any proprietary product that has a
       Frank> source component would have this problem (e.g., a web-based
       Frank> ecommerce system based on HTML templates and Perl scripts).

   No.  If it's proprietary, we can assume you can't modify it, can't
   redistribute it.

The ``can't modify it'' bit is an incorrect assumption.  If I purchase
a commercial RTOS with source code, I will expect to examine it and
customize it in various ways.  In addition, I will expect to write
board support routines and interrupt handlers for the particular
hardware that I am using.  The proprietor is likely to say that they
only support certain sorts of modifications (and even more likely to
say that the product is completely without warranty of any kind).

   But suppose Jonathan's shop contracts to adapt EROS to "Pacemaker".
   They supply the code under a dual license.

   Now, "Pacemaker II" is due out.  The downstream manufacturer _thinks_
   they know what they're doing, so they take the free-licensed code and
   make similar modifications to EROS, perhaps using a third party
   consultant even.

I guess I just don't find this case convincing.  I think the moment
the manufacturer uses different source code, the supplier of the
original source code is clearly no longer liable.  The fact that it is
a free fork of the proprietary product which the customer paid for
seems irrelevant to me.  It's different source code.

In fact, if anything, I think only a programmer would think that the
manufacturer was doing something reasonable.  I think the average
person on the street would think the manufacturer had screwed up.
It's like reading instructions that say ``these O-rings only work when
the temperature is over 60 degrees Fahrenheit'' and then using them
when it's just above freezing.

Of course anything can happen in a court.  If you're a big right
company and you get dragged before a jury on a liability case, you'll
probably lose.  That's just an unfortunate truth about our legal

Jonathan, I'm still hoping to hear why you feel the free software case
is different from the proprietary case.