Will it work? BiFeO3

That would decapitate the telephone pole. The good-ol' boys would have loved that.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

We are *not* amused.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Speak for yourself. The rest minus one ARE Amused(a bit).

I'm sorry, but you are wrong. Peter Gutman's presentations surrounding his paper showed images of the magnetic domains of disk drive tracks that had been erased, one, two, and more times - I've seen these images. Because the head doesn't always follow the exact same path, over-writing data doesn't flip all the domains that were initially written, leaving fringes that contain some original data, and can - in theory - be accessed by a sufficiently well-resourced agency.

Your argument about wasted space doesn't wash either. If the head could track exactly its previous path, the track width would be made smaller until it couldn't.

Clifford Heath

Sorry, I think I've told all my good electronic stories. Yesterday I f'ed up again. Simple heater ( --

That's just a legend. A hard drive with a history of ones and zeroes in almost all sectors in all transition-positions cannot be detail-analyzed to get a large bunch of contiguous information from any particular past state; there were old format schemes with fringe fields that could sometimes do this, but that was decades ago.

Yes, previous info leaves traces. No, the traces don't make coherent binary data of any useful sort; you might find one or two bytes, but the block-level readout, or filesystem-discovery scenarios, are all fiction.

Entropy wins, here; you really CAN lose information.

Decapitating the telephone pole would take a like more Tannerite than you'd need spread an active dose of butyl mercaptan.

--
Bill Sloman, Sydney

That is proof positive that the original data CANNOT be accessed; there's residue from two or three different eras, it's not possible to date-code any part of the residue to match up a significant bit-sequence from such remnants.

That data is never gonna reproduce a circuit diagram, faxed signature, picture, or software package. It takes too many bits to gather anything of utility.

Spoken like someone who has zero experience of cryptology.

You don't need every bit to discover quite a lot, even if you can't decode the entire message.

Clifford Heath.

Oh, 'in the voice of Lenny (from 'of mice and men')* "John is my very good friend."

George H.

*I have this loonie tunes cartoon in mind... mice.

Every track IS encrypted, because of RLL-style dependence of byte representation on the prior byte... garbling one bit in ten, with straight ASCII coding, you'd get lots of reconstructable text. But, not with flux changes on a hard drive, where every transition is dependent on correct reconstruction of the preceding data.

Basically, the data on the disk (the magnetic domains) is already compressed, and you need a LOT of it intact before you get a single plaintext character, not just most-of-eight-bits.

Catching mice? (Sylvester the cat)...

--
Cheers, 

Chris.

The inter-track gap is negative on modern drives.

no such thing.

--
  When I tried casting out nines I made a hash of it.

I am sure they did.

And often they are not.

The easy way to find your way out of paranoia and conspiracy theories is to follow the money.

If it was practical to recover data that had been overwritten on a disk, there would be companies providing that recovery as a service - there would always be customers willing to pay very large sums for such a recovery. No such companies exist - ergo, it can't be done.

If it was practical to recover a non-negligible proportion of the bits from an overwritten disk, it would mean there is extra storage capability from the same physical space. Hard disk manufacturers would exploit that to get more bits at a lower cost. Therefore, we can conclude that there is not a significant amount of redundant signal in the hard disk bits.

No, it is not entertaining - it is flogging the dead donkey well after it has turned to dust.

If there is a thread on climate, talk about the climate. If not, please resist the temptation to talk about it.

Note that I mostly agree with you about climate change (not surprisingly, since there is close to universal agreement amongst scientists on the basics). But I disagree with the way you bring it up and the way you discuss it. It is neither entertaining, informative, nor useful when it is out of context.

John Larkin wrote

Maybe a simple and effective way to 'erase' harddisks is take the disc out and put it on an induction heater. Now 2 things work for you:

1) demagnetizer like the old degausing tools used to demagnetize shadow masks in color sets. 2) heat above Curie point.

I would do that outside, no idea what sort of stuff is in those disks these days, cadmium or other ?

Cheap induction cooker gets you above 200 C, else something like my ebay thing:

Putting the whole thing on the coil / heater including 'tronics would maybe not work\ and make a mess with exploded caps...

Maybe in some cases. Usually for somebody who really knows the stuff finding an alternative way to get to the secrets is easier. But there are cases where the outcome of wars for example depend on cracking just the 'impossible'. WW2 has some nice examples of that. In short, everything was cracked, including but not limited to enigma,

I just hope you are in the US, makes me feel safer.

;-)

Congratulations on snipping the posts - that is something few people in this group seem to manage. But you have got the attributions muddled, and made it look like I wrote the paragraphs above.

Clifford Heath wrote

That is a nice publication. Also the references point to things I'v read long ago.

I like his discussion into MFM recording as I went into that myself in the eighties. Thanks.