News for hackers. Stuff that matters.
Re:Bitcoin trojan caught in the wild by gojomo on Thursday June 16, 2011 @10:43PM (#2664129)
Works until the malware scans the disk for keydata remnants from 'deleted' files (or even old swap pages).

And this guy was trying to implement this 'offline savings' strategy, but didn't completely understand the privacy lifecycle and transaction details — and thus last the keys to a $180K balance:

http://forum.bitcoin.org/index.php?topic=11104.0;all

Vanished in a poof of pure logic!

Re:Bitcoin trojan caught in the wild by gasull on Thursday June 16, 2011 @10:52PM (#2664154)
> Works until the malware scans the disk for keydata remnants from 'deleted' files

http://srm.sourceforge.net/

I don't use Windows but there's probably something similar for it.

Re:Bitcoin trojan caught in the wild by gojomo on Friday June 17, 2011 @02:33AM (#2664587)
Even 'srm' and similar tools might not work as expected on a solid-state drive with its own firmware and wear-leveling.

It's possible to protect your bitcoin keys from an arbitrarily-later malware incursion... but very hard, in ways even most power-users don't consider.

I like bitcoin. The current sharp edges and tragic mishaps are useful, for now, for learning about a new medium of exchange, which operates on a logic different from almost anything that we could easily analogize to.

If bitcoin or a successor takes off, I suspect carrying large balances will require specially hardened devices – secure VMs inside handhelds, perhaps?

And, a general desire for some recourse against instant irreversible fraudulent transfers might make the 'finalization' of certain transactions dependent on a remote secondary key approving (or failing to cancel) a payment, within a timeframe sufficient to deliver second-channel notification/confirmation.