At least now the libelous comments will be ranked below their penchant for frivolous litigation. I can't see that having any impact on their job-finding.

I can't find any accounts of anyone failing either. Generally when universities do research it gets published either way. I wonder if that means that...just a sec, someone's knocking at my dooCARRIER LOST

Has anyone ever gone to Mars or brought peace to the middle east? Surely if this has been possible for a long time it must be possible for you to point to two or three reliable articles where someone has done this.

'Freely available on Google' isn't anything like an equivalent set to 'possible', and things that are merely theoretical now may well be trivial a decade from now. Data that needs to stay secure for the long term can't depend on it being unrecoverable due to current technical limitations; that died with DES. I doubt it would be hard at all to lift data off a 30 year old drive; sure, credit card numbers from the 1970s aren't too useful now, but some things might be.

> Drive were quite a bit different in 1996.

Exactly.

It's not available as a commercial recovery service now, and I doubt it's practical with readily available consumer grade technology, but that's not relevant for long term security. It's an entirely plausible process that's just going to get easier with time, and if something still needs to be secret in thirty years it probably won't be. Security isn't always about "can't read this for now", sometimes it's about "can't read this ever."

56 bit DES keys used to be considered secure by some people because an impractical quantity of effort would be required to crack them. That's no longer the case, and it doesn't seem reasonable to assume that just because it's still hard to read faded bits from a 1996 era drive that it always will be. I bet I could pick the erased bits off the 14" 1979 era platters hanging on my wall with a magnifying glass and tweezers. ;-)

Welcome to the state of the art, circa '96.

http://www.usenix.org/publications/library/proceedings/sec96/full_papers/gutmann/index.html

Or just use 'shred', which is part of GNU coreutils.

This is incorrect and has been for a long time.

See: http://www.usenix.org/publications/library/proceedings/sec96/full_papers/gutmann/index.html

One and two are not sufficient for data that someone may be motivated to try to recover, and haven't been for quite a while. See:

http://www.usenix.org/publications/library/proceedings/sec96/gutmann.html

At the seminar where this paper was discussed, one of the comments was from someone that did steps 1-4 regularly as the only reliable way to delete important data. I'll agree 5) is a bit over the top. ;-)

It sucks how they won't heat up generic brand food too.