Sup everyone, a newbie here. I got tricked into GMing an online Shadowrun campaign, and the past five days or so I've just been absorbing as much information about the setting and rules as possible. Luckily, none of the players are familiar with this stuff either, so I'm not in too much of a disadvantage here.

But, I digress. I've got two questions to ask now, with more possibly coming later:

Firstly, I want to create an NPC hacker who's basically as much of a bastard as possible, screwing with the weaponry and implants of the players in the middle of a firefight. Partially to show the party hacker that he can do some neat stuff, too. And partially just to be a bit of a dick. Now, how would he go about most effectively doing this? I do understand the rules about hacking into their smartgun, but the signal rating's 0. Does the hacker absolutely have to be within 3 meters of the gun to pull it off? I don't perfectly understand how the entire PAN stuff works, to be quite honest.

Secondly, I intend to have the players break into a small surveillance center in the first game. Basically, the intention is that they have a roomful surveillance equipment of various sorts, storing their findings in a single secure node that the party hacker's intended to break into. How would that sort of a system be actually set up? I've designed the said secure node already, but would it logically be hidden? Would it have a very low signal rating, since it's basically there just to store and process data from nearby devices? Would that mean that the hacker would have to get physically in the room (which would be good) or would there be ways around that? Or am I thinking this in a wrong way, entirely? How would such a thing be typically set up?

EDIT: Additional question: Let's say the hacker accesses the said node in VR. What would he actually see inside something like that? Basically everything is possible in virtual realities, but what would a boring old node that's only used for data processing look like? Stereotypical green room with floating cubes representing programs and data packages? On the subject matter, what would a commlink using factory settings look like? What's the "Default" look for VR?

Also, feel free to throw any advice related to GMing in general. There are threads for that already, sure, but feel free to mention something if it's particularily important in your opinion, hacking related or not.

The VR in a node is a total virtual environment. Think of it like the holodeck in Star Trek. Anything is possible from dreamscapes to cinemascapes. I'd imagine most things come packaged with the "default windows background" style environments to chose from. Your plesant meadow, or maybe a 3d interactive rendition of Starry Night by Vincent van Gogh.

As to how an enemy hacker can get into your player's things, since they're new to the setting they're not going to be taking sensible precautions which their characters likely would. But, it's generally assumed that a runner keeps all their cyber devices slaved to their hidden-mode comlink to keep them from being jammed up or needing IC in every piece of electronics they own. That means that the way in for an enemy hacker will be through their coms. This gives them a much larger radius to be affected. And the enemy hacker can reach them remotely from whatever is the nearest wireless device he can control. So, he goes through the matrix to reach a nearby sensor bank. Using the sensor bank, he keeps tabs on the runners and tries to sniff out their hidden coms signals. Once he has identified the com signal, he'll run an exploit to try and bypass the firewall. I like to assume that all coms, like computers today, come with a little bit of shitty security. Again, this is something that your party will take care of once they get more experienced. In the mean time, maybe they've got a trial copy of Symantec master IC lvl 2. So the enemy hacker has to stealth his way around while the IC runs analyze 2 in an extended test to notice his presence. Meanwhile, the hacker can jam the players up pretty fierce by messing with their cyber, smartguns or AR imagelink overlays.

You have to be within 3 meters to pull most of the cyberware/smartgun hacking tricks, with a Signal Rating of 0. Maybe you can have a micro-transmitter (to boost the signal) with some hacking nanites (to allow backdoor access to the ware), and have it tagged on the character somehow, but the target is probably is going to notice it. I personally think trying to create an adversarial hacker in the manner that you described is a waste of your players' time. Not only are you going to waste time and resources trying to set them up for a fall, you are going to draw attention away from the run in general. Why would a single hacker be out to get the PCs anyway? Corporate spiders are more likely paid to defend a node than trying to dick over a shadowrunner's life. Maybe if the consequences of one of the shadowruns totally ruined the life of some corporate hacker, or they were going up against an underworld Yakuza or Mafia hacker.

For your first few runs, I'd suggest as FEW moving parts as possible. Get used to the flow of combat first, and make sure that every person in the team has something to do and has a chance to shine. The player sitting bored is, guaranteed, the first one to drop out of the group. Make sure that when every player creates a character, they also list several in-character goals, out-of-character goals, and "Reasons For Gaming". Knowing this ahead of time will make things go more smoothly. Sometimes, you feel that "oh, this is obvious, they want this and such and such...", but the player really wants something else. Ask them what they want... it doesn't hurt, and it is more than likely going to help.

For your surveillance center (what in the world are you doing with a "surveillance center"? Are you running a military campaign or something?), just have the peripherals (the sensors) all connected with a physical connection to a central secure node. For a first run, I would suggest a node that has a Rating 4 System and Rating 5 Firewall at most. Also make sure that there is a "second option" if/when the hacker fails. Allow the players to smash and grab the whole unit, but give them an incentive not to do so (i.e. you'll be paid less because of the fallout). Make sure the party can succeed in the first run. You are trying to get them to play, not to dissuade them from playing again.

Alright, thanks for the answers so far. Also, I'd imagine the hacker to have enough motivation to attempt to screw the PCs up when he's unarmed and the PCs burst into the room he's at. It's not like I'm creating some sort of a nemesis who constantly stalks them, it's merely self-defense. A part of the possible combat encounter.

By the time you've hacked into a single person's commlink, you're already dead. It will take a minimum of a single Complex Action to hack on the fly to get an access account (and probably more than that), followed by Complex Actions to execute actions that won't cripple all of the PCs. At most, you'd take out one person's smartlink to accidentally shoot someone else with the gun. By that time, they've shot you in the head. None of this is subtle, either... it's going to be pretty obvious that you're hacking, either in VR (you flop unconscious) or AR (you wave your hands around). If the PCs are stupid enough to let a person flop unconscious and NOT know that it's a hacker or mage, then they probably should roll more perceptive characters next time. At the very least, you'd let the PCs have a Perception test to see that the hacker is hacking.

It also goes against psychology, unless your hacker also happens to be a battle-hardened veteran who wants to fight back against the terrorists or something. Even then, in that case, he'd be looking for a gun, not going unconscious.

Well, the room has plenty of cover and several armed men to give him some support. Surely it's a good idea to at least try?

How is he going to get within 3 meters of the runners? Without getting shot?

Yup, he's probably not. In any case, it's not really an important issue, here.

Coms generally have a signal strength of 2 at minimum. Like I said, the com is the door to everything else.
And, you can operate AR with a trode net in the same manner as with gloves or any other interface.

If you're dead set on some hacker hi-jinks then have the guy run off into the next room and drop a blast door between the players and himself. Or have him surrender, lay face down with his hands on his head, meanwhile trying to break their firewalls with his mind.

derp

Great Dunkelzahm the negative on this site is amazing sometimes.

My first suggestion is remove some of the crunch. If the hacker becomes aware of the PC's as soon as they enter the building have him being the hacks then, then hand wave that he gets in or hacks their stuff. Simplest way to do it might be to just time how many rounds it takes the team from breaching the perimiter to getting into the room with the strongest kill zone or where the Hacker is. At that point he executes a command against all PC's, likely the same command. If you want to give the hacker something to do have him make an intuition and hacking test to notice processor spikes on his own commlink or network spikes on his allies.

Bottom line: Rule of cool, if it would be cool to do especially if there is a rules grounding, don't get bogged down in the nitty gritty anymore then you have to.

As to the three meters things, he'll hack the comlinks and then hack the smartlinks.

It's NOT cool as a player to get your stuff hacked, especially on your first run. This violates the "rule of cool" in more ways than I can imagine.

"Yeah, remember that first run, guys? The one where I FRAGGED YOUR HOOPS UP because you didn't know anything about hacking or the rules in general?"

"Yeah. That's why we aren't playing Shadowrun anymore."

Why do people think that when you go VR, you flop unconscious? Why then do we have the ability to act in the physical realm at a -6 dice pool modifier? We're still awake, just fully immersed in the matrix.

If the players do not skinlink their gear, then I would think the hacker could use any other equipment within 3 meters of the party member as a relay, thereby reaching their gear. (WHich is why everyone I use has skinlinks, and has their commlink in hidden mode with a good firewall and an analyze 6 (optimize 3).

Yours,
Joel

HE does not need to get within three metres of the target, he (as previously mentioned) can piggy back off a device whose Signal he can reach within 3 metres of the target. This means he has to hack the device within 3 metres of the target, and then hack the target.

Assuming your hacker has 1 meat IP and 3 Hot-sim IPs.
Step 1. Give your hacker a revolver with tracker rounds (Arsenal) with a Mesh RFID (Unwired), that will create a Signal 1 node where-ever you hit.
Step 2. When you see the PCs go on full defence (1 Complex Action), and dive behind cover (1 free action). That's your first combat turn. (This is an absolutley necessary part of the plan, otherwise your probably dead right now).
Step 3. Second Combat Turn. Pop up (Free Action). Shoot (Simple Action). Acivate Hot-sim (Free/Simple Action) (and flop down back behind cover.
Step 4. Since you have the access details of the MeshRFID you don't need to hack into it, and can just piggy back off it straight away into the targets node. Giving you 3 IPs to Hack-on-the-Fly, find the program that controls the cyberware and screw with it. And thats your third Combat Turn.

A competent SAM can drop AT LEAST 2 Goons per Combat Turn. So assuming your hacker turned up with 5 buddies, and your taking on 2-3 combat types then the PCs should be putting a pistol to the back of the Hacker's head just as he tries to hack their system.

Tl;dr A competent Hacker can screw with PCs in many varied ways, but doing it in a quick-draw situation isn't the most effective.

If you have the Hacker tracking them through the system for a while, vectoring Goon squads to trap them. You can have the hacker do all the preliminary stuff like hacking the PCs commlink before they even know they're about to be in a firefight. This means when the firefight happens the hacker can safely go straight to screwing with their systems, preferrably from somewhere safely out of blast range of any stray grenades.

As a GM you can set this up by rolling lots of dice behind your screen and laughing manically for a bit whilst the PCs are trying to decide what they'll do next.

A PC example.
Your hacker is sitting on a park bench (about 40m away) from the target you want to "extract" and his two bodyguards. They're feeding ducks. There is a small information ARO near them that describes the pond. There are kids playing on the other side of the pond.
You switch to VR, your head slumps forward and you start drooling on yourself. You look like a passed out drunk - good thing you won't stay this was for too long.
You hack into the RFID that houses the ARO, then piggy back off that to hack into the bodyguards Cybereyes and Ears.
You load a piece of Negator software (Unwired - it modifies AR so that you don't sense particular things) that negates the van your SAMs are about to pile out of, you also load descriptions of your compatriots into the Cybersafeties of the bodyguards. All set, you give your buddies the go order. The van comes through the park - the bodyguards see it as an Icecream truck and pay it no mind, thinking its heading to the kids.
The driver of the van activates a Camera Neutraliser (Arsenal) on the van, your buddies jump out of the van, the bodyguards draw their weapons and try to fire them - they don't work.
Your buddies take out* the bodyguards and grab the target. *Kill, stun etc. to taste.
You delete any records of the event and all the programs you added and edit the access logs to make it look like you weren't even there. For good measure, one of your buddies drops an EMP grenade near the bodyguards, frying their electronic devices memories.
You get up and walk away, by the time LS arrives you've blended back into the faceless mass of wageslaves.

This example assumes the bodyguards are rookies who have not slaved their Cyberware to their commlink and/or have no IC. If neither of these are true it adds more complication to the actions, and if you screw up you will alert the guards that something is up.

Keep in mind, that if you're playing by the book and your players know the rules, an NPC hacker can't do anything to them, as there are at least 3 easy and literally impossible to beat methods of becoming completely immune to hacking.
I'd advise everyone to houserule that out, because everyone being immune to hacking kinda defeats the whole point of the whole matrix rules, and I assume everyone already does, even when they don't realise their way of doing it is actually a houserule, but I still like to point that out every now and then to the "Nu-uh, the matrix rules work without any houserules! We play without any!"-crowd on DS.

Having your Characters systems immune from hacking and being immune from hacking are not the same.
Oh my you can't hack me, but wait I'm standing in a lift with a rating 3 firewall, whats this the doors won't open, the safety systems just crashed and oh look at this, the breaks just disengaged. How high up did you say I was again?

Edit: I'd actually like to know how you do that. I could probably work it out if I read the rules with a 'how do I exploit this' microscope, but I don't tend to do that. I know of at least one though: tune it off. Which brings me to the amusing point: 'we houseruled that out', wait, what, 'you houseruled you can't turn you commlinks off?'.

Alright, thanks. I've pretty much got it.

You're better off explaining the risks of a smartgun being hacked, by letting an NPC tell them a story about someone who had it happen to them. Make it a fun, memorable story. Theyll probably pick up the hint, and if they don't, they'll probably realize it's their own fault - just make sure to give fair warning.

In response to Heath Robinson:

page 237, same book:

Though Matrix initiative is concurrent with physical initiative, any actions you take using your matrix initiative can only affect other things in the Matrix. If you want to interact with the physical world, rather than the VR Matrix, you need to either go offline or concentrate on your meat body and spend a turn using physical initiative (not forgetting the -6 dice pool modifier for physical action in VR).

Emphasis mine.

That would be a rule given in order to cover what happens when you disable the override. The gear section entry has the rules for doing so, and a restating of the fact that being in VR normally prevents you from moving.

Still says nothing about being unconscious, just that your body is limp as if you were.

It's a shorthand for "falling to the floor, as if unconscious, and being unaware of their surroundings, as if unconscious". I know I'd go insane if, for the sake of being 100% correct, I had to type that out every time I wanted to discuss how obvious it is when somebody is going into a full immersion mode.

That is correct, as long as the method for immunity takes ressources or effort. The problem with the matrix rules is that they require neither. Computer systems you'd rather not want to get hacked (ie: all of them) are conceptionally immune to hacking attempts. It's as if people get to decide if they'd rather take the ballistic rating of their armor or "rating infinity" to defend against being shot.


Turning off your comlink would make 'your Characters systems immune to hacking while not being immune from hacking', as you described. Now, if everyone could just do that without it actually impacting them, you'd have an unhackable world. Fortunately, deactivating electronic devices does (unsurprisingly) negatively impact you, so that's not a method that works. Sadly, there are others. They mostly stem from the fact that the authors of different parts of the matrix rules had different conceptualizations of how the matrix was to work with no overarching vision or a proofreading to catch those incompatibilities.

First method: The rules about system design and node hierarchy tell us that the nodes of a matrix system don't need to map the physical hardware that runs it. So, a single computer can have multiple nodes while a meshed network can have just one. This is a really good thing, because it lets you have the "target system" be a single node if you want without having to design the node mazes of editions past. It heavily simplifies things by abstracting away stuff you don't care about. Sadly, the author writing the exploit rules didn't know about that change as he wrote the rules, because they actually state that you have to break into concrete nodes one at a time. So, you can have an arbitrarily high amount of nodes without any problem (which is good), while a higher number of nodes actually takes longer to subvert (which is horrible). So, you just declare your comlink to be a million nodes chained in a row and every hacker has to exploit through all of them to get to hack you. It doesn't even matter if your comlinks is unsafe and the hacker is good enough to get admin access in a single test, because he now has to spend a million complex actions before being able to do anything.
So, you have to houserule the world to either directly map nodes to physical computers (which is a band-aid at best), houserule everyone to simply not do that (which is a houserule just the same) or rewrite the hacking rules to not care about the number of hacked nodes.

Second method: Nodes are allowed to explicitly restrict access to specific access IDs. This is good, if you want to design isolated systems you have to break in and look at a workers comlink to hack, so you can get your PC-hacker onsite, but sadly nobody told the author writing the exploit rules about it. The rules let you skip the log in and enter a system without a valid username and password. Infact, you get into the system without an account at all, so no amount of wierd concrete account shenanigans can keep a hacker out, because the method he uses is abstracted away (which is a good thing, because you don't actually care about that), but there's no way to bypass the access ID test without having a valid ID. And while you can easily spoof your own ID to any number, the only way to get to know an allowed number is scanning an icon that's allowed in. Systems that don't need to allow remote admin access can thus just not have any allowed access IDs. And all those that need to have that can just route all their traffic through a thousand household items that are all only allowed in by the previous one in line, so they'd take an equally arbitrary amount of time to hack, just like method one. And the hacker doesn't get a test to bypass that, so it's unimportant how bad the system and how good the hacker is, because it's a concrete measure in a system abstracting access.
So you have to houserule the exploit test to gain access to a system to not only be able to bypass password requirements, but all requirements (which causes problems with the isolated system scenario) or rewrite the access ID rules to make sense.

I actually forgot the third method, as I haven't played by RAW matrix rules in a long time. It was equally easy to set up and impossible to beat, however.


And hell, that's a brutal wall of text. I get way too agitated by the matrix rules, seeing as most people on DS either know about their shortcomings already or just don't care...

Wow, those are insanely cheesy. Its the Matrix equivalent of 'rocks fall, everybody dies'. I'm all for gaming a system, but where is the fun in that. I don't think I'd houserule them out though - mainly because they're an easy way to go: you, Sir, need to pack up your things and leave, you are not welcome at our table.

What? Saying "You can't do that, although by RAW you would be allowed to" is a houserule...
I'd never play with those rules in effect, either, because they make the whole matrix pointless, but they are the actual rules. Not using them is houseruling, regardless of wether your houserule happens to be "I've rewritten the system" or "I've banned doing that"...

The first case, just because you can create an arbitrary number of nodes in a single system, that doesn't mean that you can daisy chain them ad infinitum. Besides, I'd think that running multiple nodes, and therefore running multiple systems, would divide up the response. Hell, I wouldn't mind allowing someone to onion layer their com node if every layer reduced the response rating by at least 1.

And, in the second case, where you're setting up a physical daisy chain, either you're going to be a mess of wires, or you're going to be running it all wireless anyway. And if you're running wireless, you can sniff out the last id in the chain just by sniffing out the signals and, in doing so, bypass all the others.
Although, I suppose you could just have a brick of cheapo comlinks, all chained together via fiber and each of them running an ultra-paranoid firewall with routed access. In fact, if i were playing a hacker, I would build a com deck just like that. Still, there's limits to how many of them you can carry.