3 ms·

We should probably assume all the major and important companies in our everyday technologies have been asked to do it. Microsoft was asked to put one at least i

by devx 13y ago

We should probably assume all the major and important companies in our everyday technologies have been asked to do it. Microsoft was asked to put one at least in Bitlocker, too: http://mashable.com/2013/09/11/fbi-microsoft-bitlocker-backdoor/ http://mashable.com/2013/09/11/fbi-microsoft-bitlocker-backd... It doesn't necessarily mean that they all agreed, but we should assume the closed source ones did it (since we can't verify now anyway), while the open source ones should be audited much more carefully from now on.

marshray 13y ago
I'm at Microsoft and I had lunch with Niels Ferguson the other day. He's still in the Windows division, overseeing the crypto implementation in Windows. His opinion expressed here http://blogs.msdn.com/b/si_team/archive/2006/03/02/542590.aspx http://blogs.msdn.com/b/si_team/archive/2006/03/02/542590.as... hasn't changed. EDIT: For some reason I can't reply to cperceiva directly at tis time. Perhaps I've commented too much in the last few hours? Anyway... It's a fair question. I don't actually know all of the checks which are in place around the security critical parts of the code, but I have been by the care, ability, and professionalism of the people I've met who work on it. It always seems to come down to having people we can trust cooperating in an environment of trust-and-verify, i.e., audit and review. My gut feeling is that the core crypto libraries, including the code securing Bitlocker encryption, are probably one of the worst places for a rogue insider to try to hide a bugdoor.
- cperciva 13y ago
  How much do you trust the least trustworthy person who has write access to your code? My biggest concern isn't that Microsoft or Apple or Red Hat might agree to include a back door in their software; but rather that individual developers might be bribed to "accidentally" introduce exploitable bugs.
nwh 13y ago
Just because it's open source doesn't mean a backdoor exists. Take a look at some of the Underhanded C Contest entrants, some of them are so perfectly evil, and plausibly deniable to boot. http://underhanded.xcott.com/?page_id=22 http://underhanded.xcott.com/?page_id=22

by devx 13y ago

marshray 13y ago

I'm at Microsoft and I had lunch with Niels Ferguson the other day. He's still in the Windows division, overseeing the crypto implementation in Windows. His opinion expressed here http://blogs.msdn.com/b/si_team/archive/2006/03/02/542590.aspx http://blogs.msdn.com/b/si_team/archive/2006/03/02/542590.as... hasn't changed. EDIT: For some reason I can't reply to cperceiva directly at tis time. Perhaps I've commented too much in the last few hours? Anyway... It's a fair question. I don't actually know all of the checks which are in place around the security critical parts of the code, but I have been by the care, ability, and professionalism of the people I've met who work on it. It always seems to come down to having people we can trust cooperating in an environment of trust-and-verify, i.e., audit and review. My gut feeling is that the core crypto libraries, including the code securing Bitlocker encryption, are probably one of the worst places for a rogue insider to try to hide a bugdoor.

cperciva 13y ago
How much do you trust the least trustworthy person who has write access to your code? My biggest concern isn't that Microsoft or Apple or Red Hat might agree to include a back door in their software; but rather that individual developers might be bribed to "accidentally" introduce exploitable bugs.

nwh 13y ago

Just because it's open source doesn't mean a backdoor exists. Take a look at some of the Underhanded C Contest entrants, some of them are so perfectly evil, and plausibly deniable to boot. http://underhanded.xcott.com/?page_id=22 http://underhanded.xcott.com/?page_id=22