[jacquesm]

You're wrong.

> if you looked close you might find the edges of bits that had been written before and weren't perfectly aligned.

GGP spoke about both recovering data that had been overwritten and re-assembling data from destroyed platters without the original drive mechanism.

[dalke]

Thanks. Yes, I should have included the "that had been written before" in my quote.

I didn't look too into the question of how to recover the contents of a hard disk with microscopy because I figured it would be possible, but expensive. Looking now, I quickly found a MS thesis at http://escholarship.org/uc/item/26g4p84b which recovered data from a disk using MFM. While the performance was poor, the author attributes that to the experimental setup.

Ahh, and http://www.dataclinic.it/magnetic-force-microscopy.htm appears to provide a commercial service to extra data from a hard disk using magnetic force microscopy.