shredding files on a flash drive

[Christopher Browne]

On Jan 25, 2008 9:54 PM, Kristian Erik Hermansen
<kristian.hermansen-Re5JQEeQqe8AvxtiuMwx3w at public.gmane.org> wrote:
> On Jan 25, 2008 1:50 PM, James Knott <james.knott-bJEeYj9oJeDQT0dZR+AlfA at public.gmane.org> wrote:
> > It would appear sections 4.1.4 and 4.2.1 cover what I was describing.
> > As I mentioned in a previous note, it would take a fair amount of
> > effort, as described in those section.  Also, I was not referring to
> > over-written or erased disks.  Simply recovering data, from drives where
> > some part of the hardware, other than the platters, has been damaged is
> > what I was referring to.  Those two sections cover it nicely.  Bottom
> > line, it is possible to recover data after the platters have been
> > removed from the drives.  Also W.R.T. the "tracking data", yes it is
> > written on every disk, unlike the old disk pack drives where you had one
> > servo head, which provided tracking for the entire drive.  The data
> > clock is also embedded in the data, which means so long as the head can
> > get a decent signal, the data can be recoverd.
>
> Right, but still, misaligned platters cannot be recovered :-)
> Correctly aligned platters do have a chance...

Your earlier posting pretty nicely described the position of "magic
machines" in the analysis of this.

Given a suitable "magic machine," I don't think the alignment issue
needs to be problem.

But I don't think it's out of line to expect "magic machines" to cost
on the order of billions of dollars.

Mind you, that doesn't mean it can't have happened.

Someone at BNR/Nortel did a "gedankenexperiment" describing how one
might build a machine that could crack DES in fairly short order,
given access to things like chip foundries.  It was imagined that the
NSA, who certainly have access to such things, might have built such a
"cracker" machine.

Then someone made it rather more real, as some folks from the EFF
actually did up a design for this, and built sample quantities of the
hardware, enough to do benchmarking to show that the task was
"honest-to-goodness feasible."

The book, complete with some circuit descriptions, has been available,
but went out of print...
http://www.oreilly.com/catalog/crackdes/

I would think it unsafe to make any grand decisions based on the
assumption that:
a) Nobody could build a suitable "magic machine," and
b) Nobody has.

There's enough "military industrial complex" money out there that it
is reasonable to expect that there are some "magic machines" out there
of one sort or another.  That doesn't mean that you or I can have
access to them, and the presence of "gedanken experiments" does not
establish the reality of there being a Real Magic Machine.

For those who are paranoid, sometimes with reason, taking paranoid
measures to destroy disk drives *isn't* irrational.  The EFF's "DES
cracker" demonstrates that what may at one time have been an
impossibly expensive technology may eventually become at least
somewhat readily available.

Thus, we don't know what may come available for cracking RSA 10 years
from now.  Using big keys is pretty wise, and we can't be certain that
progress in factoring mayn't make that useless.

We don't know but that nanotechnology may take some big steps over the
next 15 years, so that in 30 years, I might be able to get a $50
"backup recovery" unit that *can* read data off of circa-2005 disk
hardware.  That's certainly science fiction now, but many things we
have today were "wild, out there SF" 20 years ago.

> > One thing everyone has to bear in mind, is that data recovery is a lot
> > more expensive than backups.
>
> I think we can both agree on that!  Have a nice weekend :-)

Enjoy!
-- 
http://linuxfinances.info/info/linuxdistributions.html
"The definition of insanity is doing the same thing over and over and
expecting different results."  -- assortedly attributed to Albert
Einstein, Benjamin Franklin, Rita Mae Brown, and Rudyard Kipling
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists