Idiot's Guide to the Matrix 2.0 Options

[Dashifen]

No. I'm scooting past the hacking attempt because (a) I think it's pretty clear and (b) you wouldn't have been doing the reality filter test until after hacking. Reality filter only works on the node that you're hacking:

[Dashifen]

Now, on with the action.

@Everyone:
So now that Aku has broken into the system, he is presented with it's metaphor (in yellow below). More importantly, perhaps, is the fact that the three agents running in the host also get a chance to detect his presence. The Agents roll Pilot + Analyze vs. Aku's Hacking + Stealth try and detect him (p. 217). I'll throw the agents' rolls here, and Aku can reply with his rolls below:

Attacker: 2, 2, 3, 3, 3, 4, 4, 5, 5, 6 « 3 hits
Tracker: 1, 1, 1, 1, 1, 2, 2, 3, 3, 4 « Critical Glitch
Hybrid: 1, 2, 2, 3, 3, 4, 4, 4, 5, 6 « 2 hits

Since the Tracker critically glitched, I'm going to rule that it will be automatically "surprised" by Aku if, at a later time, it becomes aware of him. For now, however, there's absolutely no way that the Tracker will acquire Aku, unless Aku critically glitches as well.

As the advertising corporation's node overrides your Reality Filter's attempts to subvert it to your will there is a moment of disorienting static before its metaphor settles into your field of view. You appear to be in a long, underground tunnel. Looking behind you, it extends into the murky darkness for a hundred meters or so before you cannot make anything out. A similar sight confronts you in the opposite direction. A dull, musty odor confronts your vitual senses and the rhythmic sound of water dripping in the distance seems specifically timed to unnever you with each hidden plunk. The walls and floor are bare stone. From the ceiling you can see root systems plunging into the tunnel, illusory tentacles reaching down sometimes to only a meter or so above your iconic shoulders.


@Aku:
I'll need a Hacking + Stealth roll to combat the rolls the Agents are making against your icon appearing in their node. Also, if your Agents are coming online with you, roll their Hacking + Stealth once as well. I'll have to roll against them too as necessary.

[JustSix]

OK, I'm a little confused by the latest developments. Hopefully someone can set me straight. Aku has successfully hacked into the node without triggering an alert. Now that he's in, though, all the IC in the node get a chance to detect him and his agents? I thought that he was basically "in the clear" once inside, unless a security hacker or "patrolling" IC happens to wander by.

Do these IC count as "patrolling" since they possess Analyze?
If they do detect Aku or one of his agents, is an alert sounded -- or is that just one possible response?

I guess it just strikes me as odd that once you've successfully hacked in, all the IC in the node then get a chance at you too. Conceptually, it makes sense, I guess I just never thought about it like that...

[hobgoblin]

im guessing that the logic is that you icon/persona is allways visible, its just that the stealth program is working its butt of to void having it flagged as a intruder.

problem is this, there is no real info about how a agent/ice reacts to a persona ones you have hacked in without triggering an alarm. in theory you can go around doing just about anything unless you glitch or critical glitch, but there are no set rules for it...

[Azathfeld]

[Tiger Eyes]

I'm confused on the IC issue as well. According to pg 208, "Once a hacker has penetrated a node, he can go about his business without having to compete with security anymore - unless he has triggered an alarm. If the system is alerted to the intrusion, he will have to deal with whatever IC and other countermeasures the gamemaster throws at him."

As I've played it, once I've successfully gotten in, with either personal, security or admin account privileges, I don't have to roll any more unless I want to do something that exceeds the user account privilege I have. With a personal account - the basic account that Aku tried for - a user should be able to be in the system without having to have IC questioning him all the time.

Unless a corporation wants to have all of its staff - from the janitor ordering more TP to the secretary making travel reservations for his boss to the executive writing up a report - constantly getting attacked by IC and crashing the system... But I don't think you'd keep employees very long (or, more important, get 10 hours of work out of them in an 8 hour day :P ) if they have to stop all the time to answer to IC or get their brains scrambled.

Is my group just making it easier, or do other people play it this way? And yes, it is a chokepoint... but secretaries and sales people and clients need to get through a chokepoint just like the rest of us hackers. Otherwise how can they do their jobs?

[Jaid]

they're agents, not IC, strictly speaking.

you just load them onto the node, and tell them to perform analyse every round.

so, technically speaking, anything you find that says "IC can do this" is irrelevant. this isn't IC. it's normal agents.

note that it actually makes no sense whatsoever for anyone to be using IC under the current rules, since i see no reason why a node couldn't activate agents with equal ease, though it isn't specifically called out.

[Tiger Eyes]

Well, agents or IC, if they are running analyze constantly, all they should see from a successful hacker is - "oh, there's another user logged in. Hm. Looks like Paul Shmitz, HR analyst. Okay dokay."

If they are doing a full-on query, then every user is going to be getting stopped every 2 minutes to type in (scan in, biometric in, whatever) their passcode or verify their identity somehow. Do you know how annoying that would be to a wage slave? And what that would do to productivity? And how many times Suzie in accounting would have to take the IT department donuts to appologize for locking up the system... again? Just 'cause she entered a '*' instead of a '&' in her passcode? :wobble:

[Jaid]

there is no action involved on the part of the employee for the agent to use analyze.

and you don't type in your passcode. chances are, it's a string of numbers, letters, and maybe random symbols even, which you are not expected to memorise. you have it stored on your working computer/commlink.

it takes 1 action for the IC to check it. considering this is a chokepoint, it is an area that people should not need to go through often, if ever; they are sitting behind the chokepoint, and they start from INSIDE the chokepoint. the chokepoint is the spot between the matrix and the corporation's working area. only people that would need to regularly leave the chokepoint need worry.

it's really not that big of a hassle at all. between the three agents, assuming you don't want to allow anyone through without being scanned, you can make 3 checks per second.

i don't consider that to be what i would call a massive drain on resources. it literally takes no time at all for the person who is passing through, unless there is a massive lineup for some freakishly weird reason.

[Azathfeld]

[Aku]

ok, then, now that i'm "in", my next action would be finding this other node, but i'm not honestly sure how. I think it's a matrix perception test, but the text almost makes me think that i have to be aware of something in order to do so.

[Dashifen]

[Azathfeld]

[Dashifen]

Aku?

[Aku]

sorry, i started to type my actions out and my computer started to have bad lag playing eq and i forgot about it and restarted, but anyways...

As my reality filter fights with the host and losses, i try to blend in with the rest of the symbology, making a Hacking+stealth test.

1, 2, 2, 4, 4,4, 5, 5, 6

Scoring 3 hits

Hoping that that will be enough to avoid detection, i forego loading most of my agents, Except for battlecat, which i'll load on to the host and subscribe too. This frees up 1 payload, bringing me to 10 (since it's now running off of the servers resources, correct?) and it gets to roll pilot(4)+ stealth(6, but limited to 4 because of pilot, i think) got 8 dice.

2, 2, 4, 4, 4,6, 6, 6 again, i get 3 hits for it

Next up i'd just sorta like to lay low and see where traffic is going. I know scan is used for detecting wireless traffic, but what about traffic on/to a wired host?

[Dashifen]

Good news for Aku, with three hits to hide both himself and his agent, the system doesn't detect him. On p. 217 of SR4 it clearly status at the top of the third paragraph under Matrix Perception Tests "Your net hits determine how successful the examination is." Thus, his three hits subtracted from my three hits means I have no net hits at all and, thus, do not detect him.

[Aku]

I'm honestly now sure, since it now sees me as an "authorized" user on the node i'm not sure if the node would object to a user running something on this system or not.

[hobgoblin]

question is, can the node tell if its a agent passing by from a outside node for some reason vs a agent that got uploaded localy by a persona?

i would guess that as the agent was not spotted upon upload it can roam freely (basicly its just the same if it got in there on its own or was uploaded, the node does not "know" its there), but would have to independently hack into the next node if told to follow the owner (if not it will just stay put).

[Dashifen]

@hobgoblin
So, if the node does not "know" it's there, do you feel that the agent should count against the node's system/response ratings for running too many programs?

[hobgoblin]

well i never liked the idea of a DOS attack based on a agent storm...

but lett me redo that statement. its more like the node and its watchdog programs (think about it, a agent fielding analyze strikes me as the next step in anti-viral development) didnt see anything wrong with the new prosess that got started up (prosesses are probably starting and stopping all the time, even on a node thats supposedly high security. alltho perhaps with a lower frequency then a normal node).

they know its running but for the moment the stealth program was able to fool it into thinking its nothing wrong (kinda like how some viruses are made harder to detect because they change how they look internaly).

now for each action the agent takes, i would roll a new perception test for the roaming ice. basicly, any activity from the agent (or for that matter the persona) is like a person a guard is not entirely sure about doing something in the same room.

basicly he will continue to look the person over the shoulder as long as they are in the same room or the guard is for some reason told to leave the person alone.

one net hit and the question would be the legitimacy of the agent or persona (or maybe a treshold similar to when trying to gain access).

basicly the less time a persona and his agents stay in the chokepoint, the better...

[hobgoblin]

let my expand on my view when it comes to agents and their effect on response of a node.

first of all, i dont support the idea of a spiral of death. the number of programs only affect the effective response a hacker have available for things like calculating matrix initiative.

therefor any number of agents can be running on a node thats not being used by a hacker directly. most likely a server/office node will have personas running around, all being people that are using personal comlinks to access a shared storage and cooperation node.

in theory you could have a rating 1 node act as the host for any number of personas, each bringing and uploading any number of agents. its only if a person would try to jack directly into that node that he would experience the system "overload" as the responsiveness of the AR or VR enviroment generated for him would be cut back to deal with all the extra tasks the node have to handle.

basicly a node thats only being used as a server do not have to spend time and response generating the simsense feeds and the arrows needed for a VR or AR enviroment. therefor it can manage many more prosesses.

rember that today, the heaviest task a computer can be given thats purely a user experience, is the generation of images on the screen. hell, look at the requirements to run the fullblown windows vista eyecandy.

[Dashifen]

I agree with everything you said, including the bit about the Agent storm approach to bringing a host to its knees. However, to circumvent that either (a) agents offloaded from a persona onto a node has to, without justification that I can see, not effect the response/system of that node or (b) the node must have the opportunity to stop the offloading process before it's successful in order to protect its response/system. I lean toward (b) which allows a hacker to try to offload agents onto the node's hardware, but requires a test, IMO Hacking + Edit vs. System + Firewall, to do so. Fail the test, it doesn't work. Glitch, the node goes to alert status. Succeed and you have the opportunity to offload your agent and subsequently reduce the system's attributes.

Of course, I think any system worth the plastic it's chassis is made from is going to go on alert status when or if it's attributes are decreased. To me, this represents a significant drop in performance for the system, as a result of which it will at least inform human security types of the problem and they'll log on to figure out what's going on (resulting in more perception tests and a better chance that the hacker will be found).

This interpretation seems to be the best, IMO, because it offers the largest number of actions to the player with consequences as a result of failure and success when making those actions without resorting to a GM fiat in anyway, since the rules for Hacking and Editing are pretty much up to the GM anyway.

[hobgoblin]

kinda like that "classical" scene in hackers?:

"there is only one user on the server, but the workload is that of 20! i think we have a hacker!"

[Dashifen]

Yeah, basically. Anyway, I'll get back to the action of the tutorial tomorrow which'll give others a chance to chime in on the discussion between myself and hobgoblin (why do I always see hobogoblin when I glance at that screen name!!) are having.

[Aaron]

Not to kibitz, but I think that since Aku has user access, I imagine he'd be allowed to run programs on the node (including agents) if that is something that users of the node are allowed to do. It's possible that the node is important enough to deny the average user the ability to initiate a program, and so he'd need security or even admin access, but I think most nodes would allow agents to run at user level if a user decided to run one.

In any case, patrolling agents/IC would, no doubt, get a chance to Analyze the agent (i.e. a new icon in the node) when it starts running, to see if it's allowed, as well. This would mean that Aku still runs the risk of the agent triggering an alert if its Pilot + Stealth doesn't beat (or match, really) the patrolling agent/IC (y'know, when you think about it, IC is just an agent with a program that hurts).

Just my 2¥.