🚀 go-pugleaf

In article <40d488$16h@datamagic.com>, kanefsky@datamagic.com (Steve
Kanefsky) wrote:

> But it's better to assume the worst 
> when it comes to computer security and then prove that it's not true than 
> to assume that nothing is wrong.

Then don't put yourself on a network.  Period.

Also, don't run any code that you yourself have not personally written,
and of course this includes all the tools to build that software, and all
that hardware from the ground up, and the tools used to design the
hardware, and the hardware those tools run on, etc.  You'd better not plug
your computer into the wall, either, as someone might be sending out data
that way.  Do you have RF shielding in your computer room?  Don't leave
the room either; someone might sneak in.  Don't blink or sleep. 
Hermetically seal it, so insects can't get in and steal your data.  Etc.,
etc.

Read Ken Thompson's "Reflections on Trusting Trust" Turing award lecture,
and hopefully you'll figure out that you really are trusting a whole lot
of stuff out there, whether you explicitly acknowledge it or not.

>I'll stick with a more open system that's 
> easier to scrutinize.

The real question is:  have you bothered to scrutinize everything (from
the tools on up), or are you trusting someone else to do it?
-- 
        Nevin ":-)" Liber       nevin@CS.Arizona.EDU    (520) 293-2799
                                                office: (520) 621-1815