shredding files on a flash drive

[Kristian Erik Hermansen]

On Jan 26, 2008 4:45 AM, James Knott <james.knott-bJEeYj9oJeDQT0dZR+AlfA at public.gmane.org> wrote:
> Why do they have to be realigned?  The servo and clock information are
> embedded in the signal, on each platter.  You seem to be of the optinion
> that the data has to be aligned between platters.  This isn't so, though

Well, the tracking data is written to multiple platters at the same
time, as well as the data.  So, if you don't align the platters, you
get junk!  I still encourage anyone to prove me wrong.  Post a youtube
video of your success if you think it is possible :-)

> they may be in an operational drive.  When you read data from a working
> disk, the first thing to happen, is the head seeks the desired track.

The disk is not "working", so this seems useless :-(

> The individual tracks can be counted by observing the signal, as the
> head moves across the surface.  Then, when the controller thinks it's in
> the right position, it starts reading data, which contains the track and
> sector information, in addition to user & meta data.  This data is self
> clocking and can start anywhere around that track, independent of any
> other disk surface.  Like I said in an earlier note, it would be a good

The problem is that all of this will not happen if you misalign the
platters.  Now, let me propose you another question, OK?  Can you
extract any user data from a multi-platter disk drive, using only one
of the platters?  Is it possible?  How do you propose to do it?  So,
for instance, a disk with two platters is disassembled, platters
removed, separated, and left for you.  I take one platter and throw it
away.  I give you the other platter.  What data can you get?

> idea for you to read up on how drives actually work.  One good reference
> for computers in general, is Scott Mueller's "Upgrading and Repairing
> PCs".  This is an excellent book, that covers, in addition to many other
> things, how disk drives work.  It describes exactly what I've been
> telling you about disk operation.  By reading that, you'll find many
> errors in your argument.

Dude, I have one of the editions of Mueller's book.  I read it a long
time ago, actually, cover to cover.  If you don't believe me, then I
don't know what to tell you.  Newer editions seem to have expanded in
size, to almost 2000 pages or so.  When I read it, it was about 1400
pages.  It also included a CD.  I remember a cool video on that disc
where he opens a drive and you see it spinning live, the head seeking,
and spin-down, head parking, etc.  But I have already read that book,
as I said, so I don't see how that is going to help.  And, that book
is not very technical (in comparison to some other docs I have read).
I think that many of the other academic papers that I have read are
far more technical and useful.  In my senior year of university, I
worked on a data forensics project.  I did a lot of research into this
kind of stuff.  I even wrote a tool that I have never seen anyone else
implement.  It is a tool to recover data and timestamping information
from the Windows registry.  Did you know that the Windows registry has
timestamps?  And I can reconstruct nearly all your actions even if you
delete all others files on the system?  :-)  Take care...
-- 
Kristian Erik Hermansen
"Know something about everything and everything about something."
--
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