So let's say a team of runners has to break into a building and steal "Widget X" for example, which is highly classified and "Corp Y" has some pretty tight security measures.

We would assume that all the specs on "Widget X" would be kept on a node isolated from the general matrix, perhaps a wireless connection within just the building itself that could only be accessed from within the building

Then we could say that Security personel would have their own node that could concievable be accessed from both inside and outside the building

Then the Drones would also have their own node accessible from inside and outside the building

Then Corp Y would have another node with Matrix access for the general building use, where a hacker could get (technically) But be unable to access the node in which the Specifications for "Widget Y" are stored.

I don't know if I got my point across, but does that sound about right?

This sounds like a very reasonable set-up. You've come up with something very similar to one of my example sites (which are here if you want to compare notes). A couple of points to note on your set-up. I would not necessarily have the drone node accessible from outside the building. You might want it this way, but you also might want to limit it to inside the building to prevent a PC from hacking drones in advance. If outside access is required, you could have the drone access available via the security node. And related to that, you might want to consider in advance whether a PC could subvert the wireless connections in the building to access nodes you don't intend them to. E.g. they get in through a node that is connected to the external Matrix, and then, if they gain Admin access, they bounce from that node to your internal Widget X node. It's perfectly reasonable to rule that they aren't able to connect the nodes this way because it's not allowed at the hardware level. Similarly, consider whether the use of relays will spoil your plans for the run. I.e. a small drone with a good signal rating sneaks in and the hacker sits down the road jumped into the drone and hacks the internal network from there. Again, it's easy enough to come up with counters for this, but it's better considered in advance rather than in the middle of the game when the players can watch you trying to come up with reasons why their plan wont work to save the adventure. I've certainly been there.

But this looks good and so long as the details are well-considered, I think this will work very well. I'm kind of knackered. I'll give it some more thought when I've slept some.

-K.

Yes you can have isolated nodes. Think of a room full of computers with no matrix connection. You can have node isolated because they are hardwired in the building and there is no matrix connection or you could have isolated nodes because the room in which they are in is blocked out by wifi paint or some other reason..

Your outside/inside nodes.. explain those more in detail. If those two nodes communicate with each other its a road in... But if they are kept seperate by like wifi or jammers or "some reason" then its cool.

I think it is far easier to "black out" certain rooms in the building than to try and make the whole building secure. I would use conventional means to secure the rest of the building.
-normal matrix with camera and drones that are hackable
-lasers, pressure plates
-maglocks and biometric "gates"
-watchers, magic lines of defense

If you get through all that then you have gained access to the "black out" area's and your hacker has to be on site to hack the system and get the widget x.

I guess I could use that new Rail mounted Drone for the outside of the build...some of it inside too. It can get it's commands through the rails.

Also bare in mind that Matrix security changes drastically depending on how you choose to read the rules for Agents/IC, the way I read them IC uses the resources of the Node that it is active in so each Node's IC defenses are going to be more limited in a system that I design then in one that knasser does.

However something that both knasser and I agree on, make sure that you have a firm idea of all of the very nasty ways that the system can frag with you if you trip an alarm, a Decker should seldom get a second chance after tripping an alarm when playing in the big leagues.

That's true. It's not worth Ravor and I re-hashing our debate here. There's a thread around here somewhere that you can dig out if you want to consider our arguments in favour of one interpretation or the other. I don't know if Ravor feels there is wiggle room for either interpretation to be valid, or if he feels that the rules clearly support that Agents / IC must be running on the node they are active on is the only valid interpretation. In either case though, you need to decide how you will handle things as it will influence your system design. Briefly: my view is that an agent or IC can run on a node, e.g. a commlink and be active in another node just as a hacker's persona can. Ravor believes that for an agent to be active on a node it must have either loaded (or been loaded) onto that node, thus potentially affecting the Response of that node, OR be accompanied by a living person in the form of their persona. Did I fairly summarise your case, Ravor? Please correct me if I have it wrong, it's been a while.

How you interpret this will affect how you deploy IC in your systems.

As Ravor says, the issue with Hacking is not necessarily whether you can get in, defeat IC, etc. But whether you can get in undetected. As Ravor says, if the system knows you are there then it can make life very difficult. One such way is going on Alert, either because the node has detected you hacking it, or because IC or a legitimate user has noticed you and raised the alert to the system. The most drastic way is for the node to shut down in self-defence. That may or may not be appropriate. You obviously don't want your security node going down (and taking the camera feeds with it), but for your accounts computer, or a research server containing important plans, it's a viable response. And the last thing your hacker needs is to find that they've now got to carry a mainframe out of the building because it's refusing to talk to start up and let anyone copy data from it.

Two tips which I suspect you don't actually need, but which I'll mention because they're things that new GMs sometimes fall over, are as follows. Firstly, don't neglect the impact of multiple checks to detect the hacker. Some people take one look at the hacking rules, see that the PC has only a 20% chance of being detected when they hack in to node X and conclude that hacking is easy. They ignore that after being scanned by the fourth node, the chance of being detected has shot up to more than 50:50. Layering makes a big difference in a system based on probability. The second thing is to keep an NPC decker handy. It's perfectly legitimate for either support to be called in from an outside company if an alarm is raised, or even for a sysadmin or security consultant to be doing some late night work. So if things are going too smoothly or you want to liven things up, bring in some fresh opposition.

Oh, and don't neglect the joys of IC that's loaded with both Stealth and Track.

Hope all this is helping,

-K.

It's helping. I've been running SR4 for a while now, just wanted to see if I could get some new ideas. Got a character that REALLY wants to play a hacker. The way I see it, if it's something important the hacker shouldn't be able to just hack the building from the safety of some abandoned warehouse somewhere, the actually need to go on site to get to the good stuff. Otherwise all the cool new rules they made for the wireless systems will never come into play.

Just turn it into the scene from Mission Impossable with tom cruise crawling through air ducts, descending through laser wires, not touching the floor cuz of pressure plates and viola. hehehe. The fat slob hacker now has to be in shape and ready to eat a bullet like everyone else hehe.

Aye that's pretty much how I see Agents working knasser, it's been awhile since I read that section but I do seem to remember there being quite a bit of wiggle room in the entire Matrix chapter in general.

Still, I happen to believe that the Matrix runs smoother using my reading, but I respect your arguments and positions on the subject as well. (I'm not that keen on rehashing the old debate either, just pointing out the two different readings and let people decide for themselves.)

Mission Impossible could be fun. Might make damn hacker decide to get out of his "Steed" Drone. Made of the the player showing him that he should get a Steed for his hacker, then he goes and does it.....it's my fault I guess, and I don't really care that he actually did it, but I do think it's funny.

I don't overtly prevent a hacker from operating remotely, as a rule. After all, it would be a double standard when I so often have NPC hackers doing exactly that. But there are too many occasions when the rest of the team will be irritated with the hacker for not taking the same risks as them - when they need someone to crack a maglock, deal with an independent drone, disable a motion sensor, short out an electric fence etc. etc. PC hackers normally take on the hardware specialist role as well. But I have had a successful game with remote hacking. If you have a player with a good sense of the tactical and the leadership skills to do it, having a hacker play the role of Lt. Goreman in Aliens, co-ordinating the operation, it can be pretty fun. And tense.