From Newsgroup: alt.comp.os.windows-11
On Thu, 10/15/2026 12:27 PM, s|b wrote:
On Wed, 14 Oct 2026 15:18:52 -0700, pyotr filipivich wrote:
Yamn Remailer <noreply@mixmin.net> on Sat, 29 Aug 2026 19:39:12 +0200
typed in alt.comp.os.windows-11 the following:
My high school computer teacher told us the only way to truly delete or erase files
is to smash your computer's hard drive with a sledgehammer!
https://www.youtube.com/watch?v=jdui7v2B9DI
Alternate method is to bake it over a coal fire.
Charcoal will do as well.
<https://www.youtube.com/shorts/iHjeAlFY7bo>
(And I've been told they would still be able to get some data from these parts.)
The shredder is supposed to be good up to Top Secret, whereas
the demagnetizer design "was not overkill enough". Shredding
is a kind of physical proof. If a unit went through a
demagnetizer, and you were staring at it, the outside of the
unit gives "no evidence the procedure was carried out".
The shredder does that for you. The particulate is a certain
size. If the platters had been aluminum, just the bending
action on the shards would have been enough of a guarantee.
For glass shards from a shattered platter in a shredder,
they are unlikely to bend, they have to shatter or scratch
to release pressure applied.
For magnetic force microscopy, the sample has to be flat, the
X-Y positioner only handles 100u . You have to be able to propel
the sample smoothly while scanning via magnetic force. Presumably
piezoelectric kinds of things used on three axis. Nothing
prevents a 3.5" squared positioner from being built and placing
an entire platter under vacuum, and scanning in 100u chunks. You
would have to overlap the scans and align them, to "make" a large
image. That means the scanning process has to be *highly* reproducible.
I believe there are two scans. The first scan is AFM mode, and
this establishes the local topography (mountains and valleys),
then the AFM scan is subtracted from the MFM, so that
physical roughness does not disturb the magnetic signal extraction.
The data is default encrypted with FDE, there is a scrambler to
prevent long runs of zeros, the analog signal is PRML encoded.
Even when your scan is working, you'd be relying on CRC per sector
to prove the content was extracted correctly.
There are only a couple hundred conventional MFM in universities,
and they are used for materials research or mineralogy. That's why
a 100u scan limit, isn't a problem for them. You prepare the sample
to fit the machine, whereas in forensics, you fit the machine to
the sample. But nothing prevents the building of a different machine
for data recovery. A lot of things have to go right, and for
an aluminum platter (flatness?), this is a lot tougher than glass.
With glass though, just the damage along the edges of the shards
is a problem. The MFM head, you can't be "driving it over garbage" :-)
It's not an all terrain vehicle with a snow plow on the front.
Paul
--- Synchronet 3.21d-Linux NewsLink 1.2