Vaevictis
Aug 1 2007, 11:56 PM
QUOTE (Fix-it) |
yerr... what?
128 bit encryption is used regularly on wireless internet access (802.11x) and is ridiculously easy to crack.
your average user can do it in 10 minutes. I believe the record stands at 2-3 minutes. the more traffic, the faster it goes.
I'm sure military systems are a bit harder, but given that most hackers today aren't too interested in cracking mil systems, we'll never find out. |
That's because WEP (the encryption standard you reference) is fatally flawed, is known to be fatally flawed, and its use has been deprecated for something like 4 years, and absolutely nobody who is at all concerned about security should be using it. Anyone using WEP should have transitioned to WPA back in 2004 (or earlier!)
Get back to me when someone can break AES in 10 minutes; maybe then you'll have a point.
Of course, kzt is totally overstating the situation also; 128bit cracks will eventually be feasible, probably within 10-20 years.
QUOTE (Fix-it) |
128 bit encryption is used regularly on wireless internet access (802.11x) and is ridiculously easy to crack.
your average user can do it in 10 minutes. I believe the record stands at 2-3 minutes. the more traffic, the faster it goes. |
Not really. WEP is a 24 bit initialization vector and a 40+ bit key. The 24 bit IV is the biggest of it's several fatal flaws. It's got numerous other interesting anomalies that makes it a very flawed implementation of encryption and great example of why you should have cryptographers involved in designing encryption, and not count on the NSA to point out weaknesses of systems designed for international use. The example of WEP does point out that there lots of really crappy encryption systems and ways to screw up good systems.
The classic analysis of how WEP is fatally broken is
Unsafe at any key size; An analysis of the WEP encapsulation.
But to claim that WEP show that all encryption can be easily broken is simply wrong. AES is a decent encryption system. Please find me something that shows you can crack AES-128 using effective keys in a few minutes.
QUOTE (Vaevictis) |
Of course, kzt is totally overstating the situation also; 128bit cracks will eventually be feasible, probably within 10-20 years. |
Actually no. You run up against the laws of physics. A decent explanation is
http://sixdemonbag.org/cryptofaq.html#entropy.
You can get some interesting effects from quantum computers, if they ever work. It cuts the keyspace of symmetric encryption in half, so it turns a essentially unbreakable 128 bit encryption into a rather easy to break 64 bit encryption. This is cool, but it also turns a 256 bit cypher into an essentially unbreakable 128 bit cypher.
Apathy
Aug 2 2007, 12:37 AM
The model that the US has been pushing for 40+ years is combined arms operations. Tanks, Infantry, Artillery and Air Support all have their individual strengths and weaknesses. But combinations of all of the above allow a force to mitigate their weaknesses without significantly hurting their strengths. While I'm sure that the tools we'll be working with in 2070 will be significantly different than what we have today, I believe that this basic principle will still hold true.
So, I believe that the most effective force will have a combination of Drones/Heavy Armor/Infantry/Paranimals/Spirits/Magic, backed up by Artillery/Air Support/Ritual Magic/Spell Defense, and supported by attached ECM/ECCM. All of the pieces will make important contributions, it's just a question of how much of each type would be necessary.
Drones provide a huge amount more 'bang for the buck', but lack the the flexibility/adaptibility/intelligent decision making of humans (who are generalists by design). They can lay down a hell of a field of fire, but won't be as good at discrimitating friend from foe from neutrals, navigating unexpected terrain, 'winning hearts and minds', working independently if/when commo goes down, etc. If you're in a prepared defensive position with no civilians (Desert Wars) and controled avenues of approach, the drone-to-man ratio might be 20-to-1, but if you're conducting MOUT ops in an occupied civilian city that you need to hold with minimal infrastructure damage, the ratio might only by 1-to-1 or 2-to-1.
Cthulhudreams
Aug 2 2007, 12:45 AM
QUOTE (Apathy @ Aug 1 2007, 07:37 PM) |
The model that the US has been pushing for 40+ years is combined arms operations. Tanks, Infantry, Artillery and Air Support all have their individual strengths and weaknesses. But combinations of all of the above allow a force to mitigate their weaknesses without significantly hurting their strengths. While I'm sure that the tools we'll be working with in 2070 will be significantly different than what we have today, I believe that this basic principle will still hold true.
So, I believe that the most effective force will have a combination of Drones/Heavy Armor/Infantry/Paranimals/Spirits/Magic, backed up by Artillery/Air Support/Ritual Magic/Spell Defense, and supported by attached ECM/ECCM. All of the pieces will make important contributions, it's just a question of how much of each type would be necessary.
Drones provide a huge amount more 'bang for the buck', but lack the the flexibility/adaptibility/intelligent decision making of humans (who are generalists by design). They can lay down a hell of a field of fire, but won't be as good at discrimitating friend from foe from neutrals, navigating unexpected terrain, 'winning hearts and minds', working independently if/when commo goes down, etc. If you're in a prepared defensive position with no civilians (Desert Wars) and controled avenues of approach, the drone-to-man ratio might be 20-to-1, but if you're conducting MOUT ops in an occupied civilian city that you need to hold with minimal infrastructure damage, the ratio might only by 1-to-1 or 2-to-1. |
In MOUT, drones are not going to be laying down LMG fire left right and centre, instead you are going to be using drones as eyes in the sky, sniper style cover fire and lone style style iballs etc in house to house operations.
What you can do that would be really powerful and unique in a location with insurgents is put up a whole bunch of aerial drones with sniper rifles and shoot anyone in the streets with a gun. A powerful deterrent to carrying around your AK.
Edit: So I imagine the drone ratio is still going to be intense, like a big slow spotter flying overhead for the squad, quite a few small eyespies and lots of iball stlye drones to toss into rooms, around corners etc.
Vaevictis
Aug 2 2007, 12:54 AM
Okay. Assuming that the back of the napkin calculations there are true, I'll concede 10-20 years is possibly wrong.
But the most powerful computer permitted by physics taking a billion years? I don't see support for that claim. (Especially since I'm not quite sure that we know exactly what physics permits.)
EDIT: And beyond the notion that we don't know exactly what physics permits, we don't know exactly what math permits. Assume someone shatters the world and proves P=NP; those calculations disintegrate before the might of an algorithm created in a world where P=NP.
(Which, as I've said before, if P=NP, that would explain a lot about why Shadowrun encryption is so weak.)
Apathy
Aug 2 2007, 12:54 AM
QUOTE (Cthulhudreams) |
[/QUOTE] What you can do that would be really powerful and unique in a location with insurgents is put up a whole bunch of aerial drones with sniper rifles and shoot anyone in the streets with a gun. A powerful deterrent to carrying around your AK. |
I think this would be too arbitrary to be effective. From everything I'm hearing, it sounds like almost everybody is armed in Bagdad right now - friends, enemies, and neutrals. And not everything would be immediately recognizable as a weapon (Is that a student's backpack, or an explosive satchel?)
I absolutely agree with everything you said about using eyespies, etc for recon and as bullet soakers. I'd just worry about their judgement in ambiguous situations.
Kingmaker
Aug 2 2007, 01:02 AM
The US military does indeed have numerous drones slated to come out over the next several decades, but most are for either non-combat or very specialized roles. One is a drone cargo truck, which has the obvious utility of reducing the number of REMFs needed for each soldier. There are also smaller "mule" UGVs for resupplying troops in the field. Pretty much all drones, even the ones design for the purpose of direct combat, are being design to support real troops, not replace them.
Thyme Lost
Aug 2 2007, 01:31 AM
QUOTE (Kingmaker) |
Pretty much all drones, even the ones design for the purpose of direct combat, are being design to support real troops, not replace them. |
This is more how I see it.
Drones giving support... not taking over...
Fix-it
Aug 2 2007, 03:30 AM
QUOTE (Apathy) |
I absolutely agree with everything you said about using eyespies, etc for recon and as bullet soakers. I'd just worry about their judgement in ambiguous situations. |
that is why you have riggers sitting overwatch to provide such judgement.
WEP was a poor example, but it does prove a point:
crypto only works when done correctly.
and all the most advanced cryptographic techniques can't hide other information: eg: where your transmitter (and you) are located, who you are transmitting to, the context of the transmission, (does he report in to main hq after every attack? does he suspect that we're moving in on him? does he report in at regular intervals?)
all of this is just as important as the actual content of the messages.
QUOTE (Fix-it) |
WEP was a poor example, but it does prove a point:
crypto only works when done correctly.
|
True. And there are many ways to hose a perfectly good cryptographic system by screwing up some detail of implementation or choosing bad keys, etc. It's actually quite difficult, expensive and painful to run an end-to-end highly secure communications network. It's why very few organizations even try and the ones that succeed are typically military or intelligence organizations with a lot of money and focus on security.
QUOTE (Vaevictis) |
EDIT: And beyond the notion that we don't know exactly what physics permits, we don't know exactly what math permits. Assume someone shatters the world and proves P=NP; those calculations disintegrate before the might of an algorithm created in a world where P=NP.
(Which, as I've said before, if P=NP, that would explain a lot about why Shadowrun encryption is so weak.) |
We know what quantum mechanics and information theory tell us are the limits in terms of time of an operation and the energy cost of each operation.
And the typical symmetric encryption algorithm doesn't have the theoretical possibility of clever mathematical solution that the public key algorithms do. There isn't any clever mathematical operation that should get you the answer without a search of the entire keyspace.
So to assume that P=NP universally means there are no problems that can't be solved rapidly and correctly. This would have about as much impact on the world as the development of the printing press A minor side effect would be that it makes the police able to rapidly and perfectly solve crimes based on minimal evidence. . . .
But I wouldn't be totally shocked to find that some PhD in NSA had as his dissertation an effective factoring solution to RSA public keys in 1990. RSA and (at least) several other public key systems do depend on a set of problems that are may be susceptible to solution by really clever math. It's unlikely, but I wouldn't be shocked.
Cthulhudreams
Aug 2 2007, 06:44 AM
QUOTE (Apathy) |
[QUOTE=Cthulhudreams,Aug 1 2007, 07:45 PM] [/QUOTE] What you can do that would be really powerful and unique in a location with insurgents is put up a whole bunch of aerial drones with sniper rifles and shoot anyone in the streets with a gun. A powerful deterrent to carrying around your AK. [/QUOTE] I think this would be too arbitrary to be effective. From everything I'm hearing, it sounds like almost everybody is armed in Bagdad right now - friends, enemies, and neutrals. And not everything would be immediately recognizable as a weapon (Is that a student's backpack, or an explosive satchel?)
I absolutely agree with everything you said about using eyespies, etc for recon and as bullet soakers. I'd just worry about their judgement in ambiguous situations. |
Oh I agree. But in a situation like that I agree with fix-it, you can afford to have the drone call in a human (it doesn;t even have to be a rigger!) to make the shot, and whats more, everyone has an IFF beacon (their comm link) in shadow run.
And hell, even if you only have 6 drones up each with a jumped in rigger to prevent/control hacking, thats a pretty damn scary deterrent to walking around with that AK. I mean sure you may not get shot this time, or even next time, but it might happen this week. And thats pretty bad for you.
Even if you don;t opt for that, being able to vector in troops at pretty short notice is great, and so is the ability to provide 24/7 suvelliance coverage. The trick to defeating IEDs is to make him deploy his weapon repeatedly and 'punch air' via alternative routing etc. But unless you can find that weapon eventually an IEDer is going to get lucky, hit the route you take this week and stuff will blow up - unless you can find the him when he deploys the weapon. If every route is covered all the time by drones, he's risking exposure every time he attempts to deploy the weapon - a big deterrent
Vaevictis
Aug 2 2007, 06:59 AM
QUOTE (kzt) |
We know what quantum mechanics and information theory tell us are the limits in terms of time of an operation and the energy cost of each operation. |
Right, so we know what our current best science tells us. And as I conceded earlier, according to our current best science, maybe 10-20 years isn't going to happen. And maybe, according to our current best science, it isn't going to ever happen.
But our current best science isn't necessarily perfect, and so what our current best science says is possible is not necessarily the same thing as what really is possible.
Those figures given in the link you gave assume our current level of technological and scientific development; as I said, if there's an earth shattering breakthrough, they may no longer hold.
QUOTE (kzt) |
And the typical symmetric encryption algorithm doesn't have the theoretical possibility of clever mathematical solution that the public key algorithms do. There isn't any clever mathematical operation that should get you the answer without a search of the entire keyspace. |
... that we know of. Nobody's proven that P!=NP, hence nobody has proven that such a clever algorithm doesn't exist.
QUOTE (kzt) |
So to assume that P=NP universally means there are no problems that can't be solved rapidly and correctly. This would have about as much impact on the world as the development of the printing press A minor side effect would be that it makes the police able to rapidly and perfectly solve crimes based on minimal evidence. . . . |
According to my understanding, that is not at all what it means. There are problems that are harder than NP, so P=NP doesn't mean all problems are solvable rapidly and correctly. And it doesn't necessarily even mean NP problems would be solvable rapidly -- assume for example that the order of the polynomial is constant but large.
And I think you may be grossly underestimating the impact of the discovery that P=NP; I think the printing press would be the lower bound.
As mentioned, if you want an easy explanation for why encryption is so weak, define P=NP in the Sixth World, and your problem is solved.
In order to have encryption not work it has to be universal. There are at least a dozen different encryption systems out there that are considered highly secure and they work in all sorts of ways.
As this isn't reflected in the rules in other places (as in you can miss with a gun, as real time solutions to complex ballistic system of equations is trivial compared to universally solving encryption, and you can crash you car, as total stability and vehicle control is also trivial in comparison) I don't think this retrocon is plausible.
And I though about somewhere between discovery of fire and printing press, but I didn't want to push the comparison.
kigmatzomat
Aug 2 2007, 04:25 PM
QUOTE (Jaid) |
so just use whatever it is that makes stealth RFID tags undetectable, and therefore unhackable...
(commonly assumed to be that it doesn't broadcast until it's told to, and then only for brief periods of time, after which it shuts down, as far as i understand) |
RFID systems are passive, unpowered. The RF equivalent of an echo chamber. Normal RFID works by sending a signal, let's say "Middle C" in audio terms, and the echo "sounds" differently, maybe A-sharp, which according to the RFID definition might mean "13". Send a range of RF frequencies and get the data based on each different echo.
Secure RFID is designed to only echo if a very precise signal is sent. The only way to get a response is to keep trying each possible signal until you get a response. At a 0.001 second signal/response time, it would take a near-infinite time to find the code with any modern bit space.
You could probably break the code on a secure RFID chip with an electron microscope but at that point you can probably extract the data directly.
kigmatzomat
Aug 2 2007, 04:43 PM
QUOTE (Kingmaker @ Aug 1 2007, 05:20 PM) |
Drones in 2070 will be used heavily to support infantry and other troops. They cannot replace real soldiers. For one, even the most durable drones are complex pieces of electronics and machinery. I see a UGV breaking down far more often than a fit infantryman. Drones cannot interact with the local population, thus making occupation hell. Drones are less agile and able to operate in an urban enviroment. Drones aren't going to invent some novel way of using a weapon.
On a side note, does anyone else think that the drones in Shadowrun are extremely cheap? |
Drones are dirt cheap in SR4 but most of them are also on par with a Toys R Us remote control toy in terms of survivability.
Which is actually good for the infantry. Look, if every infantryman has an iBall drone or two (one with the smoke option, the other flashpak) then they have cheap, disposable, eyes they can chuck around as needed. If the drones include MAD and explosives detectors they can find a lot of the dangers out there, even hanging on the troops' vests.
The infantry rigger will probably fill part of the scout/point man role and part of the radioman, probably providing backup for the real commo expert. They have a backpack full of mini & micro drones, RFID-sized sensor buttons, probably some laser mics that can double as data relays for RF-free communication, and a honking powerful comm with enough signal to punch past any jammer that isn't powerful enough to make itself known as a target to the firebase.
I figure the LMG-toting drones will mostly be the province of either armor or cavalry divisions.
Shrike30
Aug 2 2007, 05:01 PM
LMG-toting drones make sense to include with a mechanized infantry group... they're not that large (you could carry several field-stripped ones in the back of a humvee equivalent), they provide surveillance and firepower, and dragging the extra equipment around to turn them into a pretty serious perimeter without exposing your soldiers or vehicles to as much as you would otherwise is easy.
With light infantry units, I can see going more the route of recon and close combat drones (like the iBall), but light infantry units do pass on a lot of the firepower available to the more conventional mechanized units simply to meet their design requirements, and recon is sort of their thing.
Dizzman
Aug 2 2007, 05:13 PM
<i>How large are the CAS/UCAS/NAN/Atzlan armies in 2070, and how are they armed/equiped/enhanced? Do you picture them as a relatively small, elite force, or are they huge hordes of mostly expendable nobodies?
My take is that they would have large numbers of expendable cannon fodder in the infantry. They would have very little in the way of personal enhancements, but would be very well equiped. (Why spend money implanting a smartlink in PVT Snuffy when it's cheaper to give him smart goggles, and when you can get the goggles back at the end of his service agreement? And if they're going back into the civilian sector, how deadly do you want them to be?)
Cyberware could then be offered to more senior soldiers as re-enlistment encentives. (Re-up for 5 years and we'll throw in your choice of Wired Reflexes or Bone Lacing!)
What do you think the equipment list would look like for a light infantry squad?</i>
Just to riff on the good stuff in this post, you'll probably have some differences based on where the nation is, who they think they're fighting and the resources they have. In general, I imagine the military will rely much more on drones and spirits. Some breakdowns.
Tech Heavy Infantry
Country Characteristics: These countries will have a strong tech base, plenty of money and a expensive, but well educated labor force. Much like the U.S., Japan or U.K. today. For missions that require lots of boots on the ground, they will probably contract out the work. The Imperial Marines would be the best example in Shadowrun.
Terrain: A high reliance on drones, tanks, etc, makes them better suited to rapid deployment in large open areas. They might not have the manpower to hold ground in large urban, mountain and jungle areas where the terrain favors lightly equipped locals who know the terrain.
Average Squad: The fighting force will probably be broken up into squads trained for fighting and tech support squads. The fighting squad is likely made up of one drone rigger with well trained and highly cybered marines supported by multiple drones. The military will probably have a long tail with a lot of technical support making up the bulk of the force.
Key Gear: Milspec armor, Military Grade Comlinks, Helmets with all the best vision and hearing enhancements, Wired 1 or 2 reflexes (Alpha), Bone Lacing (Aluminum or Titanium) (Alpha), Datajack, Control Rigs, Knowledge & Lingua softs, Sleep Regulator, Muscle Aug or Muscle Toner.
Key Skills: At least one member of each squad with good ECCM, Piloting and Technical Skills with other members trained to take over if the lead Rigger falls.
Light/Cheap Infantry
Country Characteristics: Relatively poor countries with lots of cheap, uneducated labor to draft/conscript into the infantry. And a culture that won't or can't object to you treating your infantry like sh*t. Good real world examples would be Russia and China, especially during the cold war. Azatlan would be the prime example in Shadowrun.
Average Squad: A bunch of grunts with combat drugs lead by a non-com with either some technical skills or some basic cyber.
Terrain: Quick deployments and air coverage in open areas would be their Achilles heal - drones, tanks, air support would cut them down. They would be better in jungles, urban areas and mountains where they can flood the zone with cheap troops.
Key Gear: Combat Drugs, Wired 1, Aluminum Bone Lacing, Datajack (for the virtual instruction).
Key Skills: Basic combat skills probably honed quickly by virtual instruction.
Magical Infantry
Country Characteristics: Awakened countries with lots of magical support to draw on. Companies would probably be based around a Amazonia would be the prime example in Shadowrun.
Terrain: Close quarter terrains were good tactics and their innate strength can overcome small numbers (i.e. cities, jungles, etc.).
Average Squad: Probably varies greatly from specialists squads made of adepts, shapeshifters, etc. As a force extender, they might have squads based around a non-com with magical skills (shapeshifter, free spirit, shaman, phys ad) and a number of mundane troops with basic combat gear acting as support. Bound spirits are likely attached to each squad.
Key Gear: Binding materials, foci, fetishes, magical or natural combat drugs.
Key Skills: Magical knowledge and active skills, survival skills, close and ranged combat skills.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please
click here.