Is there any data that you want to be **completely unavailable** to law enforcement with **proper warrant**?
YES. We should not attempt to bend the rules of physics or disrupt the working structures that hold our society together simply for the benefit of our nation's police forces, at any level. I don't care who they get to sign off on it, building a time machine to go back in time and snoop on any documents that have historically been un-snoopable (even if it were possible) is not the way to fight nebulous enemies of the state. The difference between an invisible time machine, and blanket surveillance of all communications "which isn't looked at until there is a warrant" is essentially the same. We wouldn't give anyone the state-sanctioned ability to go back in time and use infrared cameras to peep through bedroom windows of even Marilyn Monroe, because it's simply not ethical; if they can do it to her, they can do it to anyone. And here we are, doing it to everyone, recording the whole thing, and calling it OK because hey, it wasn't a human being behind the telescope.
Look, I don't dispute that there are bad people attempting to do bad things. The question is really one of cost, and there's a popular Benjamin Franklin quote going around I could refer you to. My own take is that, if you give up the freedoms that have made America the best country on earth, you are sacrificing the parts which most make America worth defending in the first place. It's throwing the baby out with the bathwater. Secret power is a dangerous society-killing drug: just say no.
I recall reading that spam makes up some 70% of internet traffic. Get your keywords into spam, and your noise propagation will massively skyrocket. Can you take over a botnet and repurpose it? That should be your goal, if so. If not, you might get involved with encryption of some kind. There's plenty of room for extra noise in encryption streams; throw in a few keywords into headers or tack it onto hash algorithms and you might have something as well.
I don't think you're going to get much traction with getting people to add something new to their work routine; at the scale we need, you're not even going to be above the noise floor. We already have noise generators which are of dubious effectiveness (mind you I run that one anyway).
Alternatively, do something to improve Linux usage in general. Once it becomes more widely used by Grandmas of the world, it's easier to close holes that allow the NSA to do what it does, or for knowledgeable people to write high-level versions of the kind of programs you're talking about. Think of having Tor relays on by default in more or less every neighborhood in the US. It's already a thorn in the side of snoopers; if it becomes a default option in for example Ubuntu, then the wider Linux is deployed, the greater effectiveness that change will have.
Sadly, I have little hope for change on the grassroots front. Specific projects like the Truecrypt code review and similar things no doubt happening en mass in Linux are going to be the major drivers for change as far as I can tell
I'm going to have to object strenuously to this. All software? All operating systems? You're talking about planned obsolescence - Microsoft will love the idea if they can avoid the user-fed blowback. But, you're talking about not only forcing upgrades at regular intervals, you're talking about the end of software entering public domain as its copyright expires. The copyright will obviously still expire, but if it's un-hack-ably prevented from being able to run, it might as well cease to exist altogether. And that's ignoring the utter impossibility of such draconian DRM being so bulletproof and usable before the cutoff date.
Granted, it takes a long time for copyright to expire even as it is, but what would have happened if Pong, Super Mario Bros., and other classic games that people still look back to were to just up and vanish? Maybe you don't play games, and that's OK. But to say that future generations could not look back at what was created before, and learn firsthand the joys and sorrows of early game design and see how it affects modern game development is asking digital society never to learn from its past, to say nothing of more serious applications that are created every day. Not to mention business-critical software tied to industrial machinery. The only way to make this practical is to put exceptions in the policy big enough to invalidate the whole concept.
I recall having tried Comodo some time ago, and found that it actually had more options than I wanted. Not that control is a bad thing, but going through a training process where you get interrupted every 5 minutes for a couple weeks by processes asking for permission to run is more trouble than it's worth for me. I like that feature in a firewall, but not so much an AV.
However, I do most of my security on the browser side with NoScript/NotScripts/AdBlock where most of the garbage doesn't even get onto the machine. All I really need from my AV is a red flag to wave if something should somehow get through and halt at run-time until I decide if I want it to run or not.
I don't expect my clients to deal with any of it, either. They're mostly in the "barely functional" category of technical literacy. So, relative to Avira (which works, but pops up a "Upgrade to our e-mail filter service!" message every day/boot last time I installed it), Avast is a good balance of effectiveness vs. hassle. I mean, the sandbox gets in the way once in awhile, but all you have to do is wait 15 seconds and then Avast restarts the process normally. That's within my tolerance; daily ads, particularly for my clients, are not.
This.
How does complete unification solve the basic lunacy of bad government? Federal politics is already a big-business game that only well-moneyed candidates can participate in; dissolving all internal borders would be subjecting every level to this stupidity. Not to mention, if you think it's hard as an individual to participate in e.g. elections and effect any kind of change, at least you can make some kind of difference on local/state level if you really try; if nothing else, you can vote with your feet (and sales tax dollars). Federal politics is well beyond the average citizen, a drop in the ocean really.
Federalization is exactly the wrong direction. We need to have "all other rights are reserved to the States" back in force. One size does not fit all.
I study cybernetics, in both organic and artificial neural networks. There is no real difference between organic and machine intelligence.
I think this assumption is a mistake - and a big one. Science still has not accounted for consciousness and isn't even close. Until it does, such sweeping statements are myopic at best, if applied to human beings.
Can you point me to a recent natural disaster where everyone else just shrugged it off? "More for me"
Hurricane Katrina. I shudder to think what government assistance will mean in the future with "Fusion Centers" at the heart of it all. Google that if you're not familiar. There's a lot of tin-foil nuttery, but just the basic facts that are admitted and publicly known are enough to make you stop and go "hmmm..."
You must devise a test for granting the machines the rights and responsibilities of personhood, and here it is: If they ask for rights, who are you to deny them?
This I agree with. We may find that, unfortunately, there may be many real organic humans who cannot pass this test. I suggest that we start by making it a required qualified for public office. Also, while I am not a big fan of anime, the Ghost in the Shell animated feature had a fascinating look at this idea. Recommended for forward-looking fans of gunporn and/or philosophy.
I'm just not following your logic here. You say above you worked with IETF and Daniel J Berstein, so I have good reasons to suspect you're closer and more familiar with the details of this subject, but it seems to me that fundamentally the random number generator is an important part of the encryption math, so your statement that:
What the NSA may have done is made it so your encrypted communications have two keys: yours and the NSA's. There is no evidence that it weakens the algorithm in any way, provided of course that NSA doesn't publish their private key.
While the cipher may be more or less exactly as advertised, the weakening of the RNG is still an important factor. If "the algorithm is not weakened in any way" is true, it's only in the strictest technical sense, and not how most people will define it. You then go on to say that NSA has simply made themselves another key in the generation process. This strikes me as being exactly backwards. Care to elaborate?
I switched from Avira due to constant obnoxious upgrade offers some time ago. If they've gotten better on that I might reconsider - but Avast works fine if you're willing to whitelist a process and then reboot. I mostly run into false positives with flash drives, so all it takes is unplug and replug in that case. What really pisses me off is, as you say, "yanking the keyboard away" and forcibly removing useful utilities which Norton/Mcafee tend to do regularly without the option to cancel. I've taken to setting up a Truecrypt volume that mounts read-only with all those things that tend to trigger it. Mcafee throws a tantrum, wanting to "reboot to remove the threat" every 20 seconds or so, but it gives me time to shut off the automatic protection and then get the work done.
Avast has also added update checking for 3rd-party programs (e.g. Java) which can conceivably be helpful for those users who aren't very good at keeping individual programs up to date.
MSE does have a certain minimal functionality that at least provides a bit of CYA for those users who insist they have protection they're going to load but don't seem very good at risk management...
Systems programmers are the high priests of a low cult. -- R.S. Barton