[ale] dban?? (Greg Freemyer)

[Neal Rhodes]

>> What's wrong with a ball-peen hammer applied to the center of the top
>of
>> the drive until the platter/head mechanism is permanently dished in?
>> Then a quick toss into the garbage can.
>>
>> Have even the 3 letter organizations succeeded in getting data of a
>> drive with shattered platters?

(Greg Freemyer)wrote:
>That is no where near as good as a single pass wipe.
>
>First, hard drive platters are metal.  They don't shatter, they bend.
>
>With a MFM microscope I don't think it would be a real issue at all to
>recover data from bent platters.
>
>If you have $25 see
>http://www.springerlink.com/content/408263ql11460147/
>
>Or the free summary at
>http://sansforensics.wordpress.com/2009/01/15/overwriting-hard-drive-data/
>
>That one paper is the only public doc I'm aware written in the 21st
>century that addresses the issue of recovery via laboratory
>techniques.  (ie. The Gutmann paper was from 1996 and is simply no
>longer relevant.)
>
>Greg

My apologies for not noting the original poster's intent to avoid
damage.  I generally don't dispose of hard drives until they present
failures/errors such that it's doubtful that the drive will even spin
up, let alone last through a complete surface write.  Or ten. 

We could have an interesting discussion on bending platters, and whether
the coating would bend or flake off.   Obviously the heads would be
toast and nothing would fly over it anymore. 

Perhaps another perspective is how many months one would have to spend
looking at the bent platters with a microscope to get anything useful.
Let's face it - most of our drives are filled with linux and the
zillions of files that go with it.  Even if there was a credit card
number somewhere in there somewhere in some file, or a picture of you
posing on the grassy knoll the day Kennedy was shot, do you think this
could be found and assembled?   Do you think someone without a-priori
knowledge that there was gold buried in there could be motivated to
spend enough months to reassemble enough of the inode and file
structure? 

Neal Rhodes