Chrissy Cruiser <
[email protected]> wrote:
Thanks for the response. I suppose what I have asked for on ACF before, and
I 'm not sure it exists, is someone to state with undeniable certainty that
this prog or that prog or both or several will be able to pass any known
detection and recreation of data.
I'd be leary of any such claim. The interworkings of data storage are
complex and are changing as disks change.
Unless a disk is overwrittten continually, there remains a bit pattern
in a layer from erased data. I have no clue as to the abilities of
tools and skills in data recovery. I do presume these are impressive
due to the importance to national security though.
I've only read what I can find on the subject and Guttman seems to
have seized this as an area on concentrated study and made his
findings public, although dated a bit.
Recovery is far more complex than the interworkings and thorough
erasing can make the job more tedious exponentially. As Gibson says, a
false sense of security is far worse than knowing that you have no
security at all. I buy into that mindset I guess.
Now, if you want real security, it will be on a clean machine that has
no internet access, such that no key loggers and such can be
introduced. That part is clear. Get a quality encryption program and
set up an encrypted volumn to hold all sensitive data. Use a complex
and long pass phrase. Check frequently for encryption updates. Even
the best have small flaws.
I have a book that I use to help retain it, because I'm just not that
smart. On a page number that has significance only to myself I chose
to start at a certain point and use the text, substituting certain
characters for other characters and making use of the high ascii
character set. The phrase is in the neighborhood of the maximum number
of characters the program will accept. It is not going to be brute
forced in our lifetimes. I do feel safe in saying that. The number of
possible permutations is mind boggling.
The passphrase is critical. It must be impervious to dictionary
attack, brute force attack, free from malicious programs that might
compromise it, not written on a post-it stuck to your monitor, and
long, very long.
For extra critical and valuable protection disable the swap file. The
machine will run slower, but you can be certain that bits and pieces
will not be written to random places on the disk. Quality programs try
to make certain that nothing is written to swap file, but why not be
certain? This is for a machine dedicated to security.
Then, if you feel that your baking recipes are in peril, you can wipe
the single file that holds the encrypted disk with Eraser using the
Gutmann algorithm (very slow), and maybe hit the entire disk with
other wiping algorithms if time allows.
Then, I'd offer that I would feel like I had real security in the
removal of those recipes.
Short of physical destruction of the equipment, that's my best idea in
personal security on a personal computer running Windoze.
For the nominal Windows PC that connects to the internet I don't
really think that real security is possible.
2 cents, keep the change...
BTW: I tried Eraser and it takes a long time. That's good. However, I
have a 30 gig primary partition and chose to stop Eraser, as I had
seen it in action. The program choked, leaving many large files with
illegal file names that were a PITA to remove. I had another no-name
program do this while I tried it years ago. The electricity flickered
off and I had no UPS at the time. I had 3/4's of my disk written in
filenames that were illegal and could not be removed.
