Let's discuss PAN security., Commlink Joe and Hacker Jim. Options

[Slymoon]

Yes, I'm following now. As long as the imagelink and smartlink are on the same network, they can share data.

But would the smartlink need to be on the network at all? Without a comm using DNI only, the Smartlink and Imagelink interface already. You shouldn't need either to be on a network, though they both could be.

Well, I'm not sure that the imagelink output is what you're looking for. Wouldn't that just be the number of bullets you have and a little red dot wiggling around? You might have to link up the cybereyes or eyecam or simrig or whatever. I'm not sure.

I suppose that is up for debate. How does a Smartgun work? Does the guncam actually capture the image, range and ballistic information from the weapon transfer that data to the smartlink processor? Then the processor strips the image and projects the dot + ammo count + other info?
Or does the processor forward the entire image (ghosted) to the image link? (seems a bit more confusing this way.)

I suppose you are right, thinking about it. The data from the imagelink may just be a wiggle dot + heads-up info.
hmm... though this setup wont let you shoot around a corner, so maybe it is both. Wiggle dot + info overlay by default, but you can 'turn on' the guncam image data for viewing around corners?

You can screw with the data going through a hub (see Intercepting Signals, and then the various Edit functions). However! You can't actually control a device unless you access it. I know, I know, sounds crazy, but look at the description for Controlling Devices. So, while you could send spam to someone's imagelink via their commlink, you couldn't turn them off unless you actually hack them, which you could do from their commlink (if it's part of the PAN).

Alrighty I can see this. Though the Comm is billed as a possible central control center for everything. (or maybe that is where some of my confusion is coming from, not central control, but central passthrough.)

Everything's wireless... Do you mean AR vs. VR or Fly vs. Probe? VR and Probe are easier in terms of numbers, but not always practical.

I mean, if you are out of range:
Say you are 20 meters away from a comm. I understand you can attempt a hack on the comm directly via direct wireless. ie: from my comm to his comm.
I could also decide to connect to the matrix and then find his matrix link and hack that way.

Where as if you were 2km away, I could only hack his comm via the matrix. ie: connect to the matrix, do an extensive search find his connection and hack into it.

The later was the way of old SR, though typically a decker didnt hang out online, they were actively in a node. So if you knew that node you could try to find him there, engage in cybercombat and defeat him. I don't recall that there was much in the way of hacking into his deck. (but again I houseruled that alot)


I know there has been alot of discussion regarding daisy chaining from comm to comm to comm or device to device and so on. But again that sounds really odd. As in I would have to hack comm A to get in range of Comm B to get in Range of Device C and finally to Comm D.
Daisey chaining would mean there effectively could not be a wireless dead zone. Save for active jammers in an area or the other mentioned methods.

So I tend to think more in lines with wireless connection that we have today:
IE: Micky D's has a wireless connection point, which is different from a Comm. As long as the comm is within range of that connection point and itself then the comm can access the matrix.
If Barnes and Nobles also has a connection point and it overlaps with Micky D's then I can walk from one to the other never losing my connection. (behind the scenes the comm automatically transfers its connection between the two for continuous matrix connectivity).
However, if there is a 10m gap between the wireless coverage of Mickey D's and Barnes and Nobles, then when I walk between them I would lose connection until I entered the range of either one. Regardless of whether or not I am within range of Bobs comm. and Bob is within either Mickey D's or Barnes and Nobles coverage area.

That make any sense?

(for the most part I think wireless connection nodes are so dense in SR4 that there is a relative continous connection. Though again it is through those connection points not daisy chaining comms/ devices.)

[Kyoto Kid]

...this has been a pretty interesting discussion. Actually picked up on e few things I wasn't even aware of.

The information overload I'm beginning to sense from all this and the Matrix Session thread makes feel that maybe someone else in our group needs to take over the Matrix Specialist role & I'll go back to playing the dainbramaged physad. (IMG:style_emoticons/default/silly.gif)

[Abbandon]

I think finding a node outside of your signal range is impossable. Theres no way you could sift through all the nodes in the matrix. If you were going to hack a node from outside your signal range you would have to know the commcode/access ID(website address)

Well it sounds like nodes are instructed to pass all traffic that isnt directed at them on without question. Deadzones do sound unlikely but if they say they are there why argue. You may be able to piggy back from one side of a dead zone to another but what if none of the nodes in the deadzone can reach a satellite or node that is connected to the larger matrix.

[Ravor]

And to add to all of the confusion, the one place in Fourth Edition Core that actually spells out how the Matrix 2.0 is put together doesn't make a whole lot of sense if you assume the ad-hac free-for-all Mesh network that other parts of the book sound like is being used.

One of the reasons I've decided that Matrix 2.0 is largely corp hype, the new Matrix is just the same old Matrix with a fresh coat of paint and some wireless routers jammed into the old jackpoints.

[Abbandon]

Man the BBB says sppof can be used against agents/drones/sprites. The FAQ says ANY node........

Its not a stretch to assume you can spoof any type of traffic whether its access ID's for drones/agents/sprites or passcodes for system nodes......

[Kanada Ten]

Man the BBB says spoof can be used against agents/drones/sprites. The FAQ says ANY node...

It's not a stretch to assume you can spoof any type of traffic whether its access ID's for drones/agents/sprites or passcodes for system nodes...

The FAQ is wrong. As usual.

Look at Controlling Devices (page 220): "you must first gain access to a device before you can control it." It never mentions using spoof against a node or device. In fact, if you read it a certain way, you may first have make the Matrix Perception test against target Persona, then access the node the agent is operating in before you can Spoof it. It may actually be harder to control drones and agents than devices, which is the way the rules imply with "Note that agents and drones will only take orders from their controlling persona, unless another persona spoofs an order."

By that reading, devices are easier to control, because you only have to access them (which makes sense, because devices don't have Pilots to defend against the Spoof, you just fake an account and they obey Command).

[Abbandon]

-I dont have to control it, I just have to appear like the thing that does with..SPOOF.
-Running a perception against a persona is one way of getting access ID, another is to decrypt the transmission between the persona and the drone and rip the Acess ID out of the air with Sniffer.

[Kanada Ten]

-I dont have to control it, I just have to appear like the thing that does with..SPOOF.

Just provide a quote or page number which agrees with you. From my reading: you have to have both the passcodes (or an account) and the access ID (or spoof) to control drones and agents, while you only need passcodes (or an account) to control a device. This is consistant with the rules.

-Running a perception against a persona is one way of getting access ID, another is to decrypt the transmission between the persona and the drone and rip the Acess ID out of the air with Sniffer.

Page number or quote? Track and Perception are all I can see which grant someone else's access ID, though you could make an arugement for Browse.

[Abbandon]

Ok I tried to look for page numbers and qoute the other night but i either could not find them or was smacked in the face with the opposite of what i thought and what you have been saying. So I will concede that you do indead have to MP a personna to gain an access ID.

Sniffer probably has 5 total sentences relating to what it does and they are spread throughout the book. I thought Sniffer was the program you use to actually read a file or hear a phone call. First you find a file or "phone call" with scan/browse/search, and then you "open it" with Sniffer. Atleast thats what i thought I read. pg221 Hacking and Accounts "Hackers can gain access to passcodes in several ways......sniffing traffic.......". I dont see why you couldnt sniffer out access ID's either.

Forgetting about sniffer..... so in order to gain an Access ID I have to MP a persona....whats a persona in a system??

Ok here is a scenario. You are a hacker, Bob is a bad guy. Bob drove his car to a club to meet with somebody. Hacker needs to install a program to kill the engine at the press of the button so the rest of his team can ambush the guy when and where they want. Bob's already in the club. Hacker is sitting on his Bike and is in range of BOTH Bob's and the car's nodes. Hacker wants to spoof commands to Bob's car so he can break into the node more easily. What are ALL of hacker's options?

#1 locate bobs commlink, hack bob's comm, percieve his persona to gain access ID, decrypt the signal to the car, sniffer it(for access codes?), then start spoofing?
#2 located bob's commlink, Analyze or matrix pecieve his node from the outside without hacking in for the access ID, decrypt the signal to the car, sniffer it, start spoofing.
#3 hack the car's node and install the program and log out??

[Kanada Ten]

I'm assuming Bob's node is in Hidden mode? We're also assuming Bob's car has a Pilot (otherwise: just hack the car). If all you want to do is shut it down, then you don't have to use Command, don't need to Spoof, and can just upload your Agent via Edit and subscribe it to your team's button node (once you hack the car node).

If it has a Pilot and you want it to take orders from you, locate and hack the car node (since you'll have to anyway), Trace the subscription back to Bob (grants you the Access ID), and then you can Spoof the car node, which would allow you to Command it like a drone of your own.

[edit] On passcode sniffing: wait around and sniff the car node, when Bob unlocks the door via remote, you'll have the passcode? But is it only to the door, or only to door level functions? Like a base account, where autopilot would be security?

[ixombie]

@Abbandon - I think I get it now. First of all, ignore the stuff about passcodes. It doesn't have anything to do with Spoof.

Theoretically, you could steal a passcode by sniffing someone's wireless traffic, if they sent their account passcode over wireless you could grab it, decrypt it, and it's yours. But a Passcode is not an access ID. A passcode is what you use to legitimately access something, it is not involved when you try to Spoof a command.

Spoofing a command requires the access ID of the persona you're impersonating. I don't think you need to hack to get that. What you need is to be on the same node as the persona, where you can perform a matrix perception test on it. Unfortunately, the game provides no info on what the threshold for that test is. Is an access ID obvious? Normal? Hidden? It doesn't say. So it's up in the air how easy it is to get a persona's access ID. But what's obvious is that you can do it one of two ways: go to the same node where the persona is and analyze it, or hack the persona's commlink, since the persona is also there - people are of course logged onto their own commlinks when they access other nodes. But you can't just find someone's node and run an analyze to get its access ID. You have to be in the same virtual space with their persona where you could "look" at it through the matrix. The persona is inside of the node, not broadcast randomly into the air where you an pick it up.

[Kanada Ten]

@Abbandon - I think I get it now. First of all, ignore the stuff about passcodes. It doesn't have anything to do with Spoof.

Theoretically, you could steal a passcode by sniffing someone's wireless traffic, if they sent their account passcode over wireless you could grab it, decrypt it, and it's yours. But a Passcode is not an access ID. A passcode is what you use to legitimately access something, it is not involved when you try to Spoof a command.

Spoofing a command requires the access ID of the persona you're impersonating. I don't think you need to hack to get that. What you need is to be on the same node as the persona, where you can perform a matrix perception test on it. Unfortunately, the game provides no info on what the threshold for that test is. Is an access ID obvious? Normal? Hidden? It doesn't say. So it's up in the air how easy it is to get a persona's access ID. But what's obvious is that you can do it one of two ways: go to the same node where the persona is and analyze it, or hack the persona's commlink, since the persona is also there - people are of course logged onto their own commlinks when they access other nodes. But you can't just find someone's node and run an analyze to get its access ID. You have to be in the same virtual space with their persona where you could "look" at it through the matrix. The persona is inside of the node, not broadcast randomly into the air where you an pick it up.

All Spoof does is forge Access IDs, it does not give you the ability to access (and thus control) a node, device, or agent. You cannot "beam" a Spoof out and have the receiver accept it (this would be Intercepting Wireless Signals, which is the providence of Edit and has nothing to do with Spoof). Once you have access to a node, agent or drone via hacking or passcode, you must Spoof the Access ID to have the Pilot or Agent accept the Command (devices do not require Spoofing, just access).

[edit] A successful Track will reveal the access ID, in addition to Matrix Perception tests, which is useful if Bob's comm is out of range. (though, I can't prove you can Track a subscription without the persona, so maybe that won't work).

On Matrix Perception, the number of net hits scored determines the number of answers you can gather about a persona. One net success would be enough to learn the access ID of a persona.

This post has been edited by Kanada Ten: Feb 20 2008, 04:30 AM

[Lyonheart]

It would be difficult, if not pointless to try and weave a web of commands that allow your agents to support each other without them all dogpiling onto one node and slowing it down and leaving the rest of your devices defenseless.

Well, if you game with an actual coder, expect to see a python script that does exactly that, but yeah, what he said.

It is however perfectly withing modern computer science capability's to negotiate exactly that sort of problem.

[Abbandon]

Sniffing only gives the passcode to the door of a car? lol. You cant use your commlink to send a fake spoof command to an agent/sprite/drone? You have to hack into their node first?? I dont buy that stuff at all.

[Kanada Ten]

Sniffing only gives the passcode to the door of a car? lol.

I was asking you, is the door a device attached to the car, or is it the car?

You cant use your commlink to send a fake spoof command to an agent/sprite/drone? You have to hack into their node first?? I dont buy that stuff at all.

Funny, I don't buy that you can go to a chatroom, make a Matrix Perception test and then command 900 cars to take orders from you. Spoof: "Drive to the dock at midnight if you have no passenger." *Poof* We obey because you have the Nixon Mask on.

You want to send commands without hacking? Intercept Wireless Signals, Edit.

[Abbandon]

I guess its up to the character who owns the car how many nodes his car has. If they are a freak and have a node for eeach and every thing the car can do sure. In general car = one node.

Thats why you have agents/ic and programs to prevent hancking, thats why you slave stuff, hardwire stuff, limit stuff.

I agree with your intercept,edit thing. Hmmmm rules say you have to hack+sniffer vs firewall +system to insert fake traffic. Dont really agree with that lol. Sniffer is for looking at things closely, Spoof is for trying to look like something else. Not only that but are you suppose to make that opposed sniffer test to insert fake traffic and then follow it up with the computer+edit or is that taken into account??

Spoof makes your transmission look like somebody elses. The car would have to beat you in an opposed test to tell that you are not really Bob calling from Bob's commlink.

Taking control:

Spoof: Once you have aquired an Access ID, you may issue fake commands to any agent,sprite,drone, or node if you are within the same signal range or closer.
-Vs agent/sprite/drone: Hack + Spoof vs Pilot + Firewall
-Vs a node: Hack + Spoof vs System + Firewall
A success on the hackers part means the target believes the orders came from the legit user. If the target wins it would most likely go on alert. If you are not within signal range then you will have to piggy back......

Piggy Back: If you have the Access ID but you are out of range of the agent/sprite/drone/node you wish to fake commands to then you will have to piggy back along the controlling devices link. Once you have located the controlling node you must Decrypt its wireless traffic:
-EW + Sniffer(3) threshold test.
If you succeed you are then able to insert fake commands with:
-Hack + sniffer vs X + Firewall (X = pilot for agent/sprite/drone, X = System for a node)
If you have given yourself an account on the drone/node(agents/sprites dont have accounts) then you can issue commands with a simple edit:
-Computer + Edit (simple success test or a threshold test GM's discretion)
If you botch the job and fail to issue commands from here there is only one method left. Brute Force.

Brute Force: You have two choices. Either you are within signal range of the target drone/system(agents/sprites would have to be geeked) and you can hack directly OR you are not within range of the target's signal range and you have to hack the controlling node and then issue legit orders from that node to the target(including agents, but not sprites? death to sprites!).
-hack on the fly/probe drone/system, sneek past/crush IC... , then Command+edit if you have an account or Hack+spoof if you dont

[Slymoon]

Hence my hesitation regarding fully connected fully wireless fully...

Controlling a game is quite easy as long as everyone knows and adheres to the rules/ houserules.
However, in the wireless section things are so amorphous that persons can easily agree to a certain set of rules and yet design a character/ device that has more capabilities than either the GM or player realizes or can account for.

On the flip side, the GM thinks he has the rules out clear as day. Only to piss the players off when he has an opponent that hacks their coffemaker poisoning them. Infinite possibilities lead to infinite problems...

ie: the car above, does it have one node? does it have 111 nodes? is each one protected? does that have to be clearly stated for each and every item the player has?

"yes I have a survival knife. it has 2 nodes. one just because and the other for the compass? and I have Firewall and IC 6 on both!" wtf over?

It seems the burden is on the GM to spoon feed what can and can't be done per item.
Clarify that the jacket you bought does indeed have a node and on that node a hacker could turn it into a straight jacket because this jacket happens to have have a self compressing feature and color changing and memory and vid screen and camera and computer and car and pistol and and and and and. (yes some of those are not specifically allowed by RAW but they are specifically dissallowed either.)

Do my condoms have nodes? after all maybe I want an auto sperm count after or a microbial count. What was my temperature and hers? How deep, is she fertile on and on.

I want my facial tissue to have nodes. I can tell if im sick or thats just a bit of pollen and of course the auto self destruction; no need to leave any material behind.


And *THEN* can someone spoof my condom!>!> "You mean I have Mydicksagarganzola Syndrome! Oh NOES!" Or is it Sniffer + Edit? What about sniffer + hack comm + hack condom + command IMMOLATE!

"My dicks on fire my dicks on fire!"

[Kyoto Kid]

...I dhink I'm coming down wid a node cold.... (IMG:style_emoticons/default/grinbig.gif)

[Kanada Ten]

I guess its up to the character who owns the car how many nodes his car has. If they are a freak and have a node for each and every thing the car can do sure. In general car = one node.

Thats why you have agents/ic and programs to prevent hacking, thats why you slave stuff, hardwire stuff, limit stuff.

I agree with your intercept,edit thing. You have to spoof your ID though... You could not intercept a phonecall between a bad guy and his goons and then just insert orders with an edit using your own voice.

Spoof makes your transmission look like somebody elses. The car would have to beat you in an opposed test to tell that you are not really Bob calling from Bob's commlink.

What's crazy about Intercept/Edit is that you don't need to Spoof, you're altering actual data packets or whatever, not creating new ones... The problem with it is you need data traffic to Edit, and probably need to Decrypt it too. With the way Edit is written, you can't do much on the fly like that, just censor or alter certain words and tone, etc, maybe add background voices ("I heard somebody, you've got somebody there." "What are you talking about, I'm all alone, I swear." *Bitch* "What did you say?!!").

When Bob calls his car from Bob's comm, the car says "Hi Bob, welcome back. Passcode?" When he's subscribed to it, he could have the car/drone be part of his node (mesh network) or a seperate node (tiered), but he's still logged on either way, so his commands don't need a passcode anymore; he's got his VPN port, and the car looks at his access ID as a safety precaution. You don't have a port, so if you call his car from anywhere except his comm, it'll say, "I don't know you." You Spoof the call and it fails to spot the flaws, then says "Hi Bob, weclome back. Passcode?" You hack inside, Spoof a command, and the drone obeys.

Now, I can see where people would say Spoof skips around the passcode, the rules are vague and opent to interpreation, largely because they don't have examples of everything. So, it could indeed be that you just need someone's ID to take control of their drones. That just doesn't sit well with me, that a car would be eaiser to steal than a Fizzypop soda machine.

Part of that might be how I view AR, it just makes it too simple to find people's ID. You're driving down the road on your Dodge Scoot, and a guy in a MCT MonsterZ breezes past. Take a glance at AR: a fraggin' cowboy saddled on a fraggin' T-Rex stomping down the road, complete with vibration. Grab his ID, Spoof the car with the command, "Take orders from this ID." Now, you can drive him off a cliff as needed? One test without even a chance at cybercombat? *lame*

[Abbandon]

Kanada your view is spoof can only be used from the inside of a node right? I could accept that if that turns out to be the case but let me ask this, if you are already inside a controlling node wouldnt spoofing be irrelevant at that point since you could just issue commands legitemately or atleast force commands??

They didnt give an example of a command sent to a drone or agent in that section on controlling. But wouldnt it be...
Command+ edit if your legit
Hack + edit if your not...
or more likely and this supports your case
hack+spoof....

hmmm. I think we are both right. I think we can spoof from our own commlink if we are close enough to the target, or by piggybacking, or once we are inside then hack+spoof unless we want to set up an account or some crap. I can see your point of view why cant you see my mine hehehe.

If i can make my signal look like Bob's why can't i use my commlink to spoof a command to a target agent/sprite/drone/node??

[Abbandon]

See my edit above kanada. Theres lots of opposed testing and stuff going on... there could be ic... the dude in the monsterZ would most likely not even have his wifi on if he is matrix stupid and cant secure it.


As far as node count!!! I believe the intent of the developers was to make everything have one node so that hackers wouldnt bog down gaming sessions(screw me if i can spell that right). However they also make nodes limited, a couple of programs on a node and the whole thing goes boom. Well to get around that people just started making multi-node designs like the old matrix.

I'd think a car was just a single node that controlled a whole shitload of sensors, but who know's what some dude has mechanically changed on his vehichle and what functions he added (rigger anyone?). the average slob going down the road is just going to have one node with its standard device rating. For players and all their gear device rating. I mean if you dont want to try and "sleeze" past a controller node and take out individual things you can always just go straight for the heart. Its up to the players and GM how complicated things get so you should not stress over how many nodes things have or what their attributes are.

[hobgoblin]

one quick thing about nodes, the way they are described in the book seems to indicate that a PAN can be seen as a single node, or for that matter a whole network of comlinks can be seen as a single node...

what im trying to say is that while the car may have 1001 small computers in it, it still only registers as one node, and can be ordered around as a node...