>> If anyone cares anymore, a gentleman who was on the CPR development team
>> -- or at least a beta tester (isn't everybody?) -- reported in a message
>> to this newsgroup that he experienced the very same problems as Alison,
>> INCLUDING the "OS corruption" crash, brought his concerns to the
>> attention of the CART team leaders, and was basically brushed aside with
>> words to the effect that "we can't worry about that now, we gotta get
>> this product out the door..."
>IME, Win95 is likely to decide to not restart for any number of
>reasons, most of which I've been unable to determine. Sometimes it's
>installing new software, or not shutting down cleanly, a failing disk,
>and sometimes it's apprently just because I was eating a chocolate
>pudding and watching Seinfeld in the same room as the computer.
>Is it possible the CPR contributes to OS corruption? Sure. Is it
>likely that it's an endemic problem? No. Is it likely that if Alison
>re-installed on that same machine that she'd see it again? I don't
>think so.
>What's an acceptable failure rate for OS corruption after a CPR
>install? One in 1000? How does your answer change if 1 in 1000 times
>Win95 fails to reboot without a CPR CD in the same zip code?
Jim, thanks for these comments. Here are some additional thoughts on
Win95 corruption:
Do I have solid proof that it was CART PR that caused the OS
corruption?
No. I loaded the game, ran it, exited normally, and powered off.
Next time I powered up, it wouldn't boot into Win95. I restored the
registry and configuration files from a ConfigSafe snapshot that I made
immediately before installing CPR, and everything was back to normal.
Might have been something else that did it, but I can't imagine what.
Perhaps I will do some more testing at some point, but it is *so*
unpleasant having to re-install Win95 and all applications due to a
hosed registry that I prefer to avoid exercises that put me at risk of
having to do so.
And a comment from a private email to me:
True. In reality, I regard OS corruption problems as a Microsoft issue,
not a CPR issue. The combination of the easily-corruptible yet vital
registry, old .ini files, and the habit of all kinds of applications and
system software installations over-writing existing dll's simply invites
trouble and creates a maintenance nightmare. As an old Unix junkie, I
feel that applications should *never* update system files and shared
subroutine libraries, and system patches that do should have rollout
capability and good logging. There's a very good discussion of this at:
http://www.arachnoid.com/lutusp/dll_article.html
Alison