Re: protect your system from altered or "patched" MI programs...

[Douglas Handy <dhandy1@xxxxxxxxxxxxx>]

Tom,

>They probably would if you could demonstrate that they broke code
>relying on documented interfaces.

And therein lies the secret to why I think IBM can easily win this in court,
even if it is deemed to not be a violation of the license agreement.  So for the
sake of argument, pretend IBM loses that part of a lawsuit.

What is Fast400 going to claim is illegal about the PTF?  Even if it is legal
for Fast400 to twiddle some bits in a publicly undocumented OS structure, what
is illegal about IBM changing an undocumented structure?

It has always been the case -- on every operating system I know -- that if you
rely on undocumented features that you are at the mercy of the vendor should
they decide to change their implementation.  For whatever reason.

There have been a few cases where undocumented quirks became widely used over a
period of time, and IBM succumbed to pressure to support something even though
it wasn't officially documented.  A case in point was on the S/36:  when you
used a roll/page key, the docs said you would not have the screen data returned
to you (like a CAxx key vs CFxx key in DDS).  But someone published a trick to
make it return the screen data, and lots of people wrote applications relying on
that undocumented trick.

But I'd put Fast400 in a different camp.  I can't think of a single legitimate
reason to claim IBM is not legally free to change undocumented internal
structures.

So even if IBM loses the license agreement allegation, at best you have a cat
and mouse game as IBM changes how CFINTxx is dispatched.  And you can bet the
next time it would be more complicated to thwart.

Doug