• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed



  • Powered by FeedBlitz

« CRN on the Air | Main | Dropping the Bloody Ball »

August 08, 2004

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83451db8a69e200d83456b8a369e2

Listed below are links to weblogs that reference A taste of things to come?:

» Politics from butterflyeffects
The topic of this article is nanotech, but I just loved the quote at the start: A conservative is a liberal who's just been mugged. A liberal is a conservative who's just been arrested. Anyway.... [Read More]

» Five Steps to Acceptance from TNTlog
The Centre for Irresponsible Nanotechnology takes time off from fretting about nanospam and quotes J.S. Haldane’s four steps that scientific theory goes through: 1) This is worthless nonsense; 2) This is an interesting, but perverse, point of view; 3) ... [Read More]

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Tom Craver

Chris:

Most internet abusers have the attitude that no one really gets hurt by their actions. That attitude may extend to physical things after nanotech. I'm not sure it'll extend to human life, even if we get brain backups - losing 5 years of your life is a bit different from losing 5 years of financial records. It's a lot easier to empathize with, also.


Reason

I dealt with my spam problem on Moveable Type with the MT-blacklist plugin. I'm not sure what the equivelant on typepad is, but you guys should think about moving to MT for that alone.

Brett Bellmore

I think we've learned a lot about secure computing and communications in the last decade, half of which can be summed up as, "Don't use Microsoft products.". Nanofactories can be made far more difficult to hack into, if that's a design goal right from the beginning.

Karl Gallagher

I can tell when my kid is playing with matches, and he has a reasonable guess what would happen if I caught him doing that. If ten years from now I see him typing some code into the computer and he tells me it's his homework how do I tell it's not? Especially if he's taking a PERL class. I'd just walk away muttering wistful thoughts about how readable FORTRAN was.

There's a usability issue for design software. How can we make so a bystander can have some grasp of what the designer is trying to accomplish?

Chris Phoenix, CRN


Write in Visual Basic? :-)

Chris

Karl Gallagher

VB's equivalant to FORTRAN for my purposes, as long that fancy object-oriented stuff doesn't get dragged in (this came up in a recent job interview actually . . .). So it would serve well as FORTRAN for readibility. But unless you seriously dig into the code you can't tell if it's doing smoking clover graphics or cracking the neighbor's PC. Making that transparent is going to be an interesting problem.

Janessa Ravenwood

Karl: I'm a VB programmer. You're not going to make code transparent, regardless of language. You have to dig into it. And OO isn't fancy for VB anymore, it's part and parcel of VB.NET.

Chris Phoenix, CRN


Gack! Stop! It was a joke! A pun on the word "visual", and ironic because it's a M$ product (and, as noted above, M$ products are insecure).

I don't think there'll be any way to make sure a product is completely safe, if the designer is trying to conceal danger. A diamond truss/foam composed of thin-walled tubes filled with high-pressure water is a strong and inert structural member. The same diamond tubes filled with high-pressure oxygen make an explosive. There are many, many other examples: the speed of sound in diamond is so high that a simple spring might accelerate a projectile to supersonic speeds. (I haven't calculated/checked this in detail.) So even if you could see exactly how the product was designed, you might not be able to see what it was for.

Chris

jim moore

Chris,
I liked the personal orbital kinetic kill weapon, but it uses 4 kilos of fossil fuel. I think that if you used a high altitude balloon and a computer guided rock you could get a similar effect and do your part for the environment as well. ;-)

LightOwl

I see the molecular nanotechnology challenge a bit like an arms race. Just like software developers and users are in a constant battle with computer virus writers and spammers today, so will we need to design protection technology to counter destructive misuse of molecular nanotechnology ( nanofactories ). I think the notion that we can control every single production unit in the future, is like trying to control every single gun today. It might be possible in the short term, but in the long term the prospects are laughable. I think the way Will McCarthy envisions the future in his book Bloom is quite accurate. I think we will probably need some kind of universal physical protection system protecting us from unknown threats. A system that will have to be tweaked constantly to withstand constant innovation from the bad guys. Kind of like an immune system for the world.

The security of a system can only be evaluated from the intensity of attacks.

Mark R

Sounds like your discussion is heading towards the superswarm option, that came out of Marlows Nano sci-fi book. And to be honest I cant see any other option yet. A decentralised superswarm of distributed (around the whole planet - maybe Solarsystem) nano sensors/machines that are capable of detecting and neutralizing any rogue nano event before its too late. Basically we will need this, and hopefully we put it in place before its too late. However, How do we ensure its never hacked and flawed? Its really very depressing that we ultimately come down to this option that implemented wrongly could be the demise of not just mankind but life on earth as we know it. Depressing, but as CRN points out has to be considered and explored to be avoided. ButI guess there could be some positives like preventing matter hackers from doing silly things with nano factories, or at least neutralising them before they cause much harm.

Karl Gallagher

Okay, take it from the flip side. If my kid's playing with matches out of my sight he'll probably get a lift home from a cop or fireman, followed by a lengthy chat and negative feedback.

If my kid is writing computer viruses or designing harmful nanomachines how can that be detected and brought to my attention? Ideally this would be noticed at the obnoxious-but-not-illegal stage so parents can correct the kids before they do real damage.

John B

Quoth Brett: "Nanofactories can be made far more difficult to hack into, if that's a design goal right from the beginning."

As a system administrator, my professional opinion is that there's no such thing as a secure computer, except perhaps one that's turned off, unplugged, and locked away.

Make it more difficult to break into? Certainly. Highly reccommended that you *do* make it as difficult as possible. It still won't protect you from bonehead users who have to download that funnybunny.exe game, nor developers who don't take the time to fully integrate security concerns, nor organizations which refuse to spend the time and money on security for their computer resources.

Or, to tie it back into nanotech: From the users who want that cutting-edge tech from a shady source and don't check the template for trojans, from developers who (for whatever reason) don't fully build security into their templates, or organizations which ignore the ramifications of hosting a nanofactory.

-John

Tom Craver

Chris: Kids and guns vs irresponsible adults:

In frontier times, kids were taught to be responsible around guns simply because guns had to be kept around. It wasn't a big concern. Today's rough equivalent is teaching kids to be safe around cars. A car is easily as deadly as a gun, and potentially more destructive of property. And we have ways of dealing with irresponsible drivers, though clearly we've accepted a rather horrific level of automotive carnage as "normal". I doubt that more than about 1 in a million would normally use nanotech in any fashion that is more dangerous than it is to let adults use cars.

It's that 1 in a million you need to worry about, the one that'd make missiles or rampaging robots or whatever. I'd say there are methods to minimize the spread of damaging attacks - the nano equivalent of sprinkler systems and firewalls - and ways to deter or detect 95% of such people before they try anything.

It should be possible to limit the damage to a handful of bad incidents around the world in any one year - perhaps at worst on par with current deaths/damage from natural disasters.

Brett Bellmore

I think we've been very foolish indeed about how we've gone about fighting the people who write viruses and worms, and we need to be more proactive when it comes to nasty uses of nanotech. But people don't generally start out with big crimes. You could deal with most of it by flooding the system with easy to find rigged "bad" nanotech designs and utilities, such that anyone who started down the wrong path would almost certainly end up in a pitfall before they got their hands on anything really dangerous, while the people who were just careless would be trained to be more careful in a less than mortally dangerous fashion.

Seriously, imagine for a moment that half the sites on the web were full of scripts and exploits that were just like the nasties use, except that they did nothing more than embarass you and direct you to pages on how to improve your computer's security... Wouldn't we have a lot less problems on the web?

micheal vassar

Great idea Brett. Needs lots of work before being ready for action, but great idea.

Chris Phoenix, CRN


Sprinkler systems, firewalls, arms races... These all make assumptions that offense will be more or less balanced by defense. That it'll be possible to detect and respond to an attack in a way that mitigates the attack.

We can already see areas where this isn't the case. A few ounces of material onboard is enough to bring down an airplane. There aren't any good defenses. And so we have a situation where people feel the need to be so ...cautious... that the letters BOB scribbled on an airsick bag were enough to shut down an airport.

All it takes is one mode of attack with intolerable consequences and no defense, and the whole system has to shift to prevention/preemption. And if that attack can't be detected in the planning stages either, then the system can't cope.

As technology becomes orders of magnitude more powerful, intolerable attacks will become easier to create. 9/11 cost only $500,000. We'll need some system-level changes. What? I don't know.

Tom, the biggest natural disasters kill hundreds of thousands of people. What would be the geopolitical and societal effects of a terrorist attack that killed hundreds of thousands of Americans?

Chris

Tom Craver

Chris:

Of course there are defenses. Don't fly. Fly smaller jets to minimize damage from any one attack. Xray and bomb sniffing machines to keep the explosives off the plane. Give everyone an ejection seat with parachute. These things are not cheap - so we limit their use, because there just aren't that many instances of bombs blowing up planes. If attacks were happening all the time, that would be reversed - making it more economical to implement such defenses than to lose all those people and planes. Look how different fighter jets are from commercial jet airliners.

The worst natural disasters happen in poor areas for a reason - relatively dense populations, with poor defenses (i.e. their buildings collapse easily) and poor infrastructure (escape routes, warning systems, rescue support, medical treatment afterward, etc).

This points to a number of defensive strategies. Spread populations out, in well made buildings. Provide means of independent rapid transportation and communication and production of survival/defensive needs. Set up sensor networks to spread the warning at the speed of light. Automated engineering systems to rapidly analyze threats and develop counters. Vastly improved portable medical equipment.

And of course, the best defense is to be inoffensive. If you don't want terrorists blowing your people up, don't do things that motivate terrorists. Stay out of other countries' affairs. Be friendly and cooperative, but don't form mutual defense alliances. Be prepared to defend and counter-attack, so no one thinks you are a soft target. Create as good and free a society as possible, so that other people in the world will want to emulate you, reducing the level of danger in long term.

Brett Bellmore

Unfortunately, I don't think it's possible to entirely avoid pissing off terrorists, when some of them are pissed off by the very fact that you don't instantly convert to their religion, and remake your society in the image they desire. Being inoffensive only goes so far.

But, yes, it's important to understand that a lot of our vulnerability to attack is a result of design decisions which could have gone in other directions, had we been aware at the time of the threat. There don't have to be so many vulnerable points where a small application of energy will disrupt life over a large area. You CAN design infrastructure to be survivable, instead of falling like a house of cards.

With nanotech, we can even redesign PEOPLE to be survivable.

The comments to this entry are closed.