Re: run a program in memory , not from hard
"Joseph M. Newcomer" <newcomer@flounder.com> wrote in message
news:t7tit3l4ve2gbavadhra6s6ns4lasukck4@4ax.com...
I remember back in the days when we were inventing the personal
workstation. The attitude
was "if a program locks up, you can just reboot the machine!" (This was in
1979, when we
were discussing the requirements for a personal workstation operating
system). I pointed
out that this was an unacceptably stupid idea. "Why not? It's *your*
machine!" was the
response. "Yes, and it is doing USEFUL WORK for me in the background, and
if I reboot it,
I lose all that work. Assume the goal is to keep the machine up at all
times, 100%
uptime." Now it is "I need to do useful things, so any limitation imposed
on me by the OS
is unacceptable". Never forget if you start doing these exotic things,
someone else sees
your app as a hostile attack. A bug in your code can take down every app
it has injected
itself into. If I can't keep mission-critical software up by blocking
your code, NO
MATTER WHAT YOUR OPINION OF ITS USEFULNESS MIGHT BE, then I no longer
control my machine.
You do. What I want is *control* of my machine. And that means keeping
out anything I
don't like, for any reason I feel like. If I cannot impose those
barriers, it is no
longer my machine. Or, to put it bluntly, if I am *unable* to say "No",
it's rape.
You seem to be saying more than one thought. First, your comparison of 100%
uptime vs. "exotic things" such as user customization of programs using
binary modification (e.g. hooks) is not valid. 100% uptime can hardly be
considered a "limitation imposed on you by the OS". We all want 100%
uptime, and there is no downside to that.
No one is arguing you should be able to say "No" and if you can't, it's
rape. What you forget is: if you say "Yes" it's consensual and should not
only be possible but promoted. As the author of hook dll's yourself, I
don't see how your position can be any different.
-- David