Linus Torvalds Admits He's Been Asked To Insert Backdoor Into Linux 576
darthcamaro writes "At the Linuxcon conference in New Orleans today, Linus Torvalds joined fellow kernel developers in answering a barrage of questions about Linux development. One question he was asked was whether a government agency had ever asked about inserting a back-door into Linux. Torvalds responded 'no' while shaking his head 'yes,' as the audience broke into spontaneous laughter. Torvalds also admitted that while he as a full life outside of Linux he couldn't imagine his life without it. 'I don't see any project coming along being more interesting to me than Linux,' Torvalds said. 'I couldn't imagine filling the void in my life if I didn't have Linux.'"
Re:Shaking? (Score:5, Informative)
Re: Would probably be found (Score:3, Informative)
As Thompson explains in his Reflections on trusting Trust (http://cm.bell-labs.com/who/ken/trust.html) even if you download everything in source form, and review it, you are still susceptible to manipulation if you use the compiler binary and haven't reviewed it's source.
Or the source of the compiler compiling that compiler, and so on.
Re:Would probably be found (Score:5, Informative)
Re:Would probably be found (Score:5, Informative)
You are doing something illegal - everyone is. You may not even know what you are doing that is illegal, but if the NSA knows everything you do, they know what you are doing that is illegal.
They aren't going to do anything about it until you do some thing that is legal that they don't want you to do.
If you run for office, they own you.
Re: Would probably be found (Score:2, Informative)
Schneier's 2006 post about a counter https://www.schneier.com/blog/archives/2006/01/countering_trus.html should be mentioned.
Look, first, to the boot time device drivers (Score:5, Informative)
Worrying about compromise of the Linux or Windows kernel is foolish - they're so large, they could have anything hidden inside and you'd never find it (searching for such is literally uncomputable). Begin your concerns with the device drivers from who knows where that are put into place by your motherboard BIOS or EFI boot systems. Conventional operating systems are entirely dependent on them, and they're completely beyond your ability to inspect or trust. And the Open Source variations have the same issue as the operating systems - large, monolithic blocks of code impenetrable to analysis.
You fear what you know about. Fear, instead, what you don't.
Re:Would probably be found (Score:5, Informative)
test subjects must be faulty if they don't immediately believe the academic's interpretation of some data presented to them.
Probably the actual discovery in this experiment: There were a lot of faulty test subjects.
Actually the similar studies have been repeated numerous times with the same result, so it is unlikely to be a fault of the subjects or the methodology. What the tests do show is that information that we hold to be technical types of information we are readily willing to concede that we could be wrong. Information that we hold as a belief or ideological position, we hold on to vehemently. Technical issues responds to logic. Ideological ones are usually emotionally based and processed in a different part of the brain. Most social views including politics and religion fall into the ideological camp and is why it is very difficult to get people to change their position using logic. It's also why, things like prejudice and bigotry are so hard to eradicate, because they, too are ideological positions.
The old adage used to be to not discuss politics or religion when having company. The tests just confirm what we already knew.
Re:Would probably be found (Score:2, Informative)
"Standardized testing" does not show any difference between races that can be attributed to genetics. What it usually shows is disparities between cultures and socioeconomic groups. The problem with people who make the claim you do is that they extrapolate actual genetic differences into explaining any sort of racist thing they want to explain. And yes, "Asian people are good at math" is just as racist as "black people are the least intelligent of the races". Obviously genetic differences exist, but they're far, far more minor than people like you would have us believe and are just matters of different tradeoffs in gene expression due to historical environmental factors.