If "production networks" cannot be rendered totally secure, they should not exist. Moreover, if they do exist they should be wholly insulated from the Internet

There's always a connection to the Internet. Sometimes it is sneakernet, sometimes it uses photonic information dellivery to bio-ocular scanning device, which uses cranial data storage and processing, and meatfingers to transmit the data through an array of buttons commonly called a "keyboard"; but there is always a connection. Hacking airgapped networks (which are still networks, just with some strange hops through biochemical computers) is just another stop on the path. If we can trick a computer into accepting a "dangerous" value, we can do the same for humans. If we can train humans to reject those values, we can train computers to do the same.

Humans are just another kind of programmable machine on the network we call Earth, with different kinds of exploitable flaws. Right now we trust the machines more than we should so their security is weaker than the humans in many cases, and so the machines are the targets. But that will change though hard experience.

Not trying to contradict you, just noodling on the nature of being a node on a network.

Susan Schneider, a professor of philosophy at the University of Connecticut, joins a handful of astronomers, including Seth Shostak, director of NASA's Search for Extraterrestrial Intelligence, NASA Astrobiologist Paul Davies, and Library of Congress Chair in Astrobiology Stephen Dick in espousing the view that the dominant intelligence in the cosmos is probably artificial.

You know, my mechanical engineer friend had some really good suggestions about the appendix surgery I was planning to get. Perhaps I should let him make the call instead of the surgeon. Oh, wait, no, that would be stupid.

Notice how there aren't any artificial intelligence researchers on that list? They are no more qualified to discuss artificial intelligence than a mechanical engineer is to discuss surgery. Better than my dog, to be sure, but not good enough to take their word for it.

I am an artficial intelligence researcher. We are cyborgs, ever more tightly coupled to the increasingly intelligent machines -- like our smart phones -- that house ever more of our memory, our social circles, and our emotional artifacts. Whatever it is that makes us who we are, increasingly, is coupled to our machines. And we will continue to be cyborgs, with an increasing share of our consciousness handed off to the machines onto which we smear our selves.

It will not be us versus them. We are them.

*Abbott -- I swear I get that wrong every time I type it.

If the job still gets done it's a good thing that jobs gets replaced by AI.
The flaw isn't in who does the work, but how the economic system around it is set up.

This is dead on the money. The traditional example is moving shoe manufacturing to China. The model says that if we move shoe manufacturing to lower wage countries, then US GDP will increase. As a result, the average income in the US will increase, even if the shoe makers in the US cannot find new jobs.

"But what about those shoe makers?"

"Well," the mathematical model says, "even if we have to provide financial aid to put those former shoe makers into jobs for which they are currently underqualified, the net economic benefit of moving the manufacturing overseas is a win for the US (and for China)."

It's actually all quite true. The mathematical model is as well-tested as gravity. But there's the rub -- right now we're just straight up shifting the cashflow out of labor and into capital gains. From those who work for a living to those who have investment money to put at risk -- without a commensurate job retraining program or economic incentives for employers who migrate those labor resources into the new economy. Done that way, it's absolute shit for the laborers. But what's worse is that leaves them as a wasteful drag on the economy instead of developing them as a productive economic resource. In the long run, it is worse even for the wealthy who are doing a little better in the short run. It is, from a purely objective economic standpoint, fiscally stupid.

And not only are we allowing the shift to happen, we're encouraging it by having a lower capital gains tax rate than the labor tax rate (complicated math, it's higher than the 15% or 20% that the left claims, and lower than the 40% counting corp tax that the right claims, but the real tax incidence of capital gains on the investor is substantially lower than the real tax incidence of income tax on a laborer with the same income).

We are creating the exact sort of economic conditions that have sparked most of the major economic revolutions since the dawn of civilization. And we're seeing the same rise of nationalist rhetoric fueled oligarchy that was at the center of each of those previous examples -- with one major change: This time it's happening in multiple countries at once. Abbot, Harper, and Cameron are as deeply tied to Wall Street and the surveillance industrial complex as Obama and the rest of the party-line Republocrats are.

I hardly doubt that a future employer would hold him accountable for telling the truth under oath.

Was that intentional, Freudian slip, or mistake? I mean, I concur 100% -- there is no doubt in my mind that the most successful US companies strongly favor a willingness to lie under oauth -- but then I've worked on Madison Ave and my brother worked on Wall Street, so I've seen the sausage get made.

detect how intoxicated you were in the photo and suggest that you not post it. Which in the end, is probably for the best.

Not bad. Now if we can just get them to suggest that you not post things when you don't look intoxicated, they'll have covered all the cases where not posting things to Facebook probably for the best.

how can we quantify the effect of simply having *some form* of security...?

Brief aside; of course we should have some security. I'm only saying that the things TSA has done are generally both unnecessary and ineffective.

quantify the effect of ... security to deter the less-suicidal ones?

The way to measure the deterrent effect of a system is by looking at the risk in cases where that system is not in place. In the case of TSA, we can look at cases where the TSA has no deterrent effect and there isn't an analog agency or system. By looking at the probability of attacks that are not deterred by TSA, and comparing that to the probability of attacks in cases where TSA is in place, we can approximate the risk mitigation.

So, for example, we have little or nothing like TSA to deter toxic gas attacks in crowded public spaces. AFAIK, the recent (and possibly accidental) chlorine gas release at the furry convention is the only case since the creation of TSA.

Consider how much terrorism we experienced in the recent years prior to TSA, the level of terrorism in first world nations that don't have something like TSA, and the number of events in the US in areas that are not protected by TSA or a similar deterrent. Compare that to the three minor terrorist attempts that made it through TSA's watch, and their threat level. Even if you take the most pro-TSA estimates you reasonably could, I think we're talking about a deterrent effect that falls somewhere below the life saving benefit of "Don't Run Near The Pool" signs -- at a much higher cost in both dollars and liberty.

is the TSA right to be cautious or have its actions caused unnecessary hassle for passengers?

The TSA has done about ten billion screenings since its inception. They have caught zero terrorists. They have missed three. All three failed, for reasons completely unrelated to TSA. TSA screenings are ineffective and unnecessary. This has been apparent for years, this story is just one more bit of security theater. TSA panders to the terror that is the terrorists' only weapon when we should be fighting it.

Seriously, what is the NSA going to do when the consequences of their arrogance propagate fully through our information culture?

One thing they'll do is get their oligarch friends to deny services to people who use encryption to keep the government from knowing their identities, like they've been doing with banks and TOR, by implying that people who use privacy protecting encryption are criminals.

Ahhh, how delightfully naive we were. Here's the tinfoil hat pessimist's prediction, from that discussion:

More likely, Apple will release a iPod update with COOL NEW FEATURES L@@K which oh yeah, btw, breaks compatibility with real-purchased songs. So then your iPod will not play your Real purchased library, until Real reverse-engineers it again, and who knows how long that'd take. So you'd have perhaps hundreds of dollars of songs on your iPod that you couldn't get to for an indefinite period of time; and Apple would just shrug their shoulders when you complain.

Silly boy, they won't just stop playing competitors' music, they'll burn the crops and sow the fields with salt! Errr, got a little overexcited there. They'll delete the files!

Grumble grumble. I came into the comments section to see reactionary histrionics and all you can manage is reasonable and dispassionate analysis?

Diversity is a good thing. I understand that, with increasing use of Linux as a desktop OS by people who don't run servers, systemd makes a lot of sense for some people.

I am the primary admin on servers in three different states. The benefits of using init for remote admin outweigh the simplicity and user-friendliness of systemd on my laptop.

I switched from Mandrake to Debian almost fifteen years ago when I first started doing heavy remote admin, I'll make a change again now, and the world will keep on spinning. Having both approaches is a good thing.

Every good law has counterpoints. Traffic signals prevent me from driving through the intersection even when there are no other cars there. Assault laws mean you can't punch someone who talks on their phone at the movies. The right to a trial means we can't just execute people we know are guilty.

One of the other examples I've been hearing lately is about Citizen's United. They say overturning it or passing contradictory legislation could hamper Steven Colbert, or limit the ACLU or EFF. Well, yes, it might. But that would be better, overall, than what we have now.

The goal is not to have laws that capture every nuance. Government is a blunt weapon that must operate in a non-discriminatory fashion. Special cases exist that show the friction in every law. The objective is not for every special case to be efficient, but for the law overall to be efficient.

Last mile providers colluding with incumbents to provide preferential access to consumers harms competition in content. Competition is good in the long run, even for the things we like that may appear to be harmed in the short run. There are natural limitations to competition on carriage, we should not extend those competition limitations to making discriminatory deals with content providers.

TL;DR version: Register.co.uk is a serial clickbaiting site, they admit it, and this article is an intentional, blatant misrepresentation of the research. Link to El Reg only for the same sort of reasons you would link to The National Enquirer.

Some have pointed out the explicit invocation of the slippery slope, but it is worse than that.

His comments to the House of Commons came after the parliamentary intelligence and security committee concluded that the brutal murder of Rigby could have been prevented if an internet company had passed on an online exchange in which one of the killers expressed "in the most graphic terms" his intention to carry out an Islamist jihadi attack.

This is not the same as blocking access to child porn sites. He is calling for the content of all packets to be inspected for unapproved speech.