mtp
|
|
response 25 of 29:
| 
Apr 26 06:47 UTC 1992 |
I understand now. I also agree that the most secure approach to the problem
would be to implement a hardware lock. But as you've pointed out - simple
locks would need to be reset physically by a technician. Additionally,
you have the problem of a wide variety of products to be protected:
notebooks, 286s, 386s, 486s. So a simple BIOS swap would be ruled out
because it would be difficult to keep up with the diferrent chipsets
that are constantly thrown at you. Cost would also be a factor - no
matter how you aquire the BIOS (in house programming or custom AMI, Award,
or Pheonix BIOSs) your cost would be too much time and money. I might as well
kiss my profit good-bye.
As far as your 50% effectiveness estimate regarding software security. I
believe you over estimate the powers of most end users. There are more
users who don't know what a BIOS or a partition is than those who do. Even if
they or someone they know *DO* know about these things, there are enough
pitfalls that can be included in the design to foil most attempts to defeat
the security. It would also be possible for me to know when such an effort
had been made (when the user calls with a down computer). The system I have
in place now has not been broken by any one of several users.
Which brings me back to my original request. If anyone can help me do this
"CHAIN" or "CALL" maneuver within an MS DOS 5.0 config.sys file, I can be
free to use this operating system. I currently require DR DOS 6.0 to round
out the security features I've implemented.
|
mcnally
|
|
response 26 of 29:
|
May 1 05:16 UTC 1992 |
What happens with your password scheme if the person just leaves
the PC on all the time? Does the password expire and the disk
turn into a pumpkin at the stroke of midnight? What if they're in
the middle of a write? But if not, your scheme isn't terribly effective
until the next time they reboot, is it?
|
