Hi.
I just read Wireless world for the second time and I can not get one thing.

What is difference between device and node?

What about security camera? Can I hack every camera? Or I have to hack some security node only?
Is it sufficient to hack riggers commlink to control his drones? Or have I to hack every drone?

I have just mess in my head

What I think (and please do correct me, read this just yesterday):

A node runs programs in AR/VR, a device is merely present inside its node. So if you get "legitimate" access to the security node, you could simply access the cameras, otherwise you would need to hack each. Your choice. I do hope comlinks count as device in this regard. Otherwise we would need to install agents with offensive utilities - otherwise known as ice - on every runners gear.

A device is any electronic item that interfaces with the matrix, and that´s just about all of them. Any device can be hacked, using the device rating in place of all matrix attributes, ECCEPT if it is part of a network (such as a PAN). If it´s part of a network, you have to hack the central node, then use the Command program on the device from inside.

This is fairly well covered in the book. The real problem is the lack of rules for most matrix actions. What kind of test is it to command surveillance cameras or turn off electrical fences? The program descriptions tells you it´s propably a command+hacking or a command+computer test, depending on your account status. But is it an extended test? Is it an opposed test? Is it a threshold test, and if so what´s the threshold? These things are up to the GM for now.

Under the headline "Linking and subscribing" on page 212, it says you can configure your devices to only interface with your commlink. Nowere in the chapter is any method of getting around this, as far as I know.

(A problem I never thought of before arises here. Nothing really prevents you from having devices configured in this way even if they are not actively subscribed. How can you hack them then? They are inaccessable to all devices eccept the owner´s commlink, but that commlink does not communicate with them. Then they´re hack-proof, right?)

I might house-rule some way of spoofing the signal of the parent device if any problematic situation occurs. No system like this should be fool proof, but the easiest way of handeling it must be to just say there might be a way of hacking the drone directly, but it would be more difficult than finding and hacking the commlink anyway.

Well to hack anyone's drone/device that is subscribed to thier PAN you still need to hack the comlink the DOMAIN CONTROLLER in other words, then using your spoof program/complex form/sprite you would "fool" the device in thinking it was a legitimate command from its PAN/Controller. This is detailed in Hacking section in BBB Wireless World. Sorry I don't have a page reference I do not have my book at work today.

yeah, what was said. if you can find the drone in the wireless, even if they're far away from their controller, you can hack into the drone and try to trick it into thinking the controller gave it the command. That's the whole thing to hacking a node. you gain entry, then manipulate it. It's just easier to trick the computer into giving the command sometimes.

at least, I think that's how it is. >.< I need to review matrix chapter again. Some devices you have to get to through another source, especially cyberware and other items on a person, because those are often linked through the neural network, not the wifi. So you'd have to get into something like a commlink that happens to also be connected to do it.

Check out the "Spoof Command" on p.224 in the BBB. When you spoof a command you send a signal that appears to be from the owner's comm so the drone/device obeys.

Maybe I am just missing one sentence, few words or something

As Kigmat said, you use the spoof command. Spoofing literally makes the drone think that said commlink sent the command. It's a fake command with a fake datatrail meant to 'counterfeit' a legit command.

When you configure a drone to only talk to a specific Comm, it does this by identifying the comm by certain properties. Spoof makes the drone believe that you *are* that specific Comm so it should obey. Most of the time you will have to hack the controller's Comm, fortunately you only have to gain access, not actually do anything.

Example:
Bob has his drone driving up and down the street, giving Mary a migraine. Mary hates Boband his annoying techno-toys so she plans on wrecking his drone. Bob's comm is in hidden mode so first Mary has to locate it (Electronics Warfare + Scan (4)). Mary finds Bob has encryption so she cracks that (Decrypt + Response (Encryption Rating x2)) and can then begin intercepting the communications (Electronics Warfare + Sniffer (3)). She tries to make changes to the data stream (Computer + Edit) but the drone is registered to Bob's Comm and ignores her changes as cruft.

She's going to have to try and masquerade as Bob for this to work, which means
Mary needs to get a good "look" at his Persona. Being a foresighted spite-queen, she's already probed her way into his Comm (Extended Hacking + Exploit) to get his remote-access password. (Bob doesn't know it yet but 30 gallons of prime, frozen bull semen suitable for artificial insemination are currently arriving at his place of work. All the records will agree that Bob's comm placed the order.) She logs in and examines his Persona (Computer + Analyze).

Mary fires up her disguise software (Hacking + Spoof) and orders the drone to aim straight for a bus three blocks away. Bob begins cursing as his new toy is crushed flat.

Mary giggles and decides if now is the time to load the spam-relaying software on Bob's comm. Bob will need the 0.0001Y/spam earned to buy a new drone. Though the fines for spamming on the corporate grounds will be far, far more.

Mary has more dark plans for Bob so she clears the records of her actions on Bob's Comm (hacking + Edit). She does a data search (Data Search + Browse) to make sure she hasn't missed something and realizes he has other drones online. She makes a few changes to the subscriber lists (Hacking + Edit) so the drones are no longer exlusively tied to Bob's comm. Hopefully that punk kid Jeph down the street will notice and cause some sloppy mischief, diverting any blame away from her.

Nice one!
Thats what I needed. Thank you a lot.

Two things I left out is that Bob's Comm gets a chance to recognized Mary's hacked login and if Bob is full VR on his Comm he might have Analyze running, which could detect Mary's virtual presence.

couldnt she just use spoof? and forget the analyse bit (Computer + Analyze).that she did before-hand?

That would be like impersonating someone you have never seen, but everyone else has. it doesn't work. You need some idea of what you're imitating for it to be capable. Mind you, in the case of making something slaved to a system fall for it, it's relatively easier (as the system is the friggin world in a space, and usually preset to an extent). A persona/rigger/hacker/whatever is trickier by nature of being unique and not being a fat entity in a digital world (you can't be present in a persona like you can a server).


*edit* actually, now that I go read the spoof command, that might not matter in a server. as long as you can convince the device that you should be paid attention to, you don't need to 'spoof' to use it (although the system will notice if you don't hide yourself). It's when you need through someone else's ID or maybe a fake one (if you can fake for those open to general use) that you spoof.

One more thing.

Example #3 on page 210 - so that drone had not been registered to someones commlink at the time, right? If it would be, it can not be hacked. But you can find the commlinks node, hack it and than spoof some nice command (like play basketball with yourself ) or unregister it.

Can you crash some drone programs from commlink node?
Or just move to drones node from commlink node?

All drones are nodes; think of their brain being a comm running a Pilot program.

A drone registered to a specific comm will only accept commands from that comm. Note that "commands" refer as much to the Command program as anything else.

The drone is still a wireless node with all the good and bad that brings, meaning it can be hacked. It isn't efficient to hack individual drones when you could hack the controlling Comm and crash/steal the whole drone net. That's why Bitsy considers it a "brute force" approach.

In my example Mary could have hacked Bob's drone but it would have been a "fast" hack (probing takes hours) that could have set off an alert, making Bob realize it was an active attack rather than a defective drone. Even if she had stealthed into the drone, she still couldn't give the drone orders until she changed its registration (Data Search followed by Edit). Then she'd have to try and quickly delete all signs of her presence on the drone in case the memory survived (meaning access logs as well as the registration of her comm ID). Way too tricky to pull off in the few seconds of time available.

She can't spoof the drone even though she's logged into it because she doesn't know what the registered controlling comm "looks" like. Modern technology today keys that work like the puzzle lockets. Each one is incomplete and only when you put them together do they make sense. This means that even if you have one of the keys (the drone) you still can't guess what the other one looks like (the comm). Future tech will be as capable, if not moreso.

Her options are to crash the OS, kill the Pilot, possibly load a new Pilot/Agent, and edit the drone's registration so she can give the Pilot orders.

Arguably she could load up a Command program while hacked into the Drone and use it to control the drone's autonomous subsystems like weapons or sensors. The Pilot is probably the only application that drive the drone, though.

depends if he ran 'security patch' or not.

As stated somewhere, using default security setting in Windoze (pun intended) is worth nothing unless you update it, preferably with non-Win software.

Meaning that Bob mignt not have change anything since Mary looked it up.
But, yes, it IS a GM call.... as much other things in the matrix section.

Actually, software upgrades aren't necessary. Many current security systems are time-sensitive, meaning the security key is only valid for a set period of time before you get a new key or that the security key updates itself using a time-sensitive algorithm and the initial handshake data.

I figure it should be based on the result of the Matrix Perception test; more successes more time. It hasn't been enough of a problem for us for me to try and hack up a house rule yet.

As I understand it its kinda like the first Deus Ex computer game. You can hack the individual camera, or you could hack the central security computer and deal with the cameras from there; the big difference is that in SR4 you do both with the same tool (commlink and hacking skill), rather than a Multitool for the camera and the Hacking skill for the computer (as was in Deus Ex).

Some wise poster (I forget who) once posted here that a node was "Something that you have to enter one password to gain access to". The book definitions say that nodes can be either a device or a network, but then go on to define network as a set of interconnected devices.

So how do you know when to treat the individual device as a node, and when to treat the network of devices as one big node? It comes down to how the network designer set things up.

It's perfectly valid to have each seperate device be it's own node, requiring its own password to log on to. Under this kind of scheme, even if one camera on the network were compromised, the rest would still be secure. The drawback is having to upgrade the firewall on each individual device, and having to enter a seperate password each time you access a new device.

The alternative is to set up the entire network to be a single node. The firewall of the central device is the only one that counts, but when it gets compromised, every device on the network is equally vulnerable.

Would it make sense to say that it is only commands that are made impossible by the subscriber list? That you could actually hack a drone or a smartlink to crash it or just examine it, but you could not spoof commands without having percieved the persona that controls it?

Does everyone agree with kingmatzomat that you could in fact hack subscribed drones, edit the subscription so that you control it, and then go to work? (Did I get you right 'matszo?) This would be devestating for the original owner of the drone.

I thought about it some and I don´t see why you shouldn´t be able to do this. The trick then off course would be that riggers usually monitor their networks closely. Also the entery would have to be brute force, because that´s required in the kind of situations were you would like to do this kind of thing. So it would propably end up being a hacker vs. hacker combat before the editing is done.

I think that's right. Actually, I appear to have thought that was the case several posts ago.

Stating the obvious or repeating past statements for clarification are parts of conversing . I just wanted to put what you seemed to suggest in my own words, and I was also wondering if there was any objections from other posters. It seems to be a rather liberal interpretation of the rules. Off course, these rules encurage and even require liberal interpretation so that´s all well and good. Still this is something I would like to hear from different points of view if possible.