#1: Under what circumstances do nodes have detectable pathways to
other nodes? I think a controlling commlink would appear as a room
with doors that lead to its peripherals (e.g., guns, vehicles,
drones), and if those peripherals are capable of running programs, and
one of the programs being run by that node is Stealth, then the door
to it will be hidden. Do doors also appear to any other nodes I'm
logged into? Or is that just one of the things you can find out by directly
analyzing my persona?

#2: I know drones can run programs, and rigger-adapted vehicles count
as drones and so theoretically could run progs, but can guns? Can normal
vehicles? What are their device ratings considered to be? Does the GM
just pick out some factory-issue commlink from the core rules that seems
commensurate with the priceyness of the vehicle, or is there a standard
default set of values in one of the supplements?

#3: If an intruder enters a node containing a guarding persona (a
hacker, or an agent like an IC) the guard gets a Matrix perception
check to detect him. What kinds of things might the hacker
do that would give that guard another such check? Does the guard
also get this chance when the hacker isn't in yet but is in the process
of trying to hack in (at the same time the node is getting its
Firewall + Analyze checks to detect the hacker)?

#4: Why didn't I just take the blue pill?

#1 I think a threshold 1 Computer + Analyze test should be sufficient to identify what you listed.

#2 See the table for the default device ratings. Typically 3 or 4.

#3 Depends on the guard, which means it's up to the GM. There may be defined things the guard is looking for, such as browse requests of a particular type.

#4. Peer pressure.

#1: Based on Pynk's response, dont worry about it unless the number of dice is less than 4 (which any low level hacker or commlink is capable of). Remember that with 4 dice, you can trade that for 1 automatic success.

#2. I think it is 4.

#3. IMHO-Any time there is a hacker vs hacker situation like this, I would assume that the defending Persona would scan each combat round or each time an intruding hacker interacts with the systems.

Hmm.. although in the case of hidden doors (nodes running Stealth) it's an opposed test against the node's Firewall + Stealth, right? As a GM I might shoot the odds a little more closely and estimate 1 success per 3 dice to oppose a PC's perception test -- or maybe make one roll for all the Stealth 3 devices if they're all probably rating 3-4 (and therefore Firewall 3-4), another one for the Stealth 5 nodes, etc.

So unless the Stealth program rating happens to be lower than the device (System) rating, we can assume the Stealth prog will be run at the System rating, and the rule is therefore: stealth dice=device rating x2.

Do things "subscribed" to my commlink, like smartlinked guns and drones, present obvious pathways that show up automatically if those devices aren't running Stealth? Would such obvious pathways exist to other commlinks I happen to be logged into, like say those of my teammates? Especially since those commlinks can't be running Stealth (and therefore, if I read the rules correctly, necessarily be in hidden mode) or else their users will appear suspicious in a public place. I guess that's why every runner needs a "front" commlink to avert suspicion and a "real" commlink to talk to smartguns, etc.



Yeah, that seems reasonable. I guess there's no canonical answer to, "Does the guard.." (who might be a PC hacker, remember) "..have a chance to detect the hacker trying to get in?" But we know I can have my commlink run Analyze on its own, using my Computer skill rating, to detect people who have already made it in. If it can also detect hackers knocking on the door (i.e. hacking on the fly) it seems like having Analyze help detect them would be standard practice and would therefore be mentioned in the Hacking on the Fly section. But maybe not, since the guarding persona would have to be logged into that node and there's a limit on how many such nodes he can be logged into at once -- so maybe the assumption was that it's usually only the node's firewall trying to detect the attempt. It just seems like it would bear mentioning that any Johnson whose comm you might try to hack will almost certainly be logged into his own node and in possession of a copy of the common-use Analyze prog, and be scanning for such things -- giving him another roll to catch incoming hackers. Or is that considered to be an unusually savvy maneuver for some reason?

Anyway, thanks much for the help, guys.

In SR4A Matrix Section, p. 228 it discusses Matrix perception Tests. Basically for each hit on the computer + Analyze you ask one question about the following.

• Access ID of a user or independent agent
• Alert status of node
• Edit date of a file
• Function of a control icon
• Hidden access to another node
• Matrix damage taken by icon
• Presence of a data bomb
• Programs run by persona or agent
• Rating of one Matrix attribute
• Type (file, user, agent, program type, etc.)
• Whether a file or node is encrypted
• Whether a Trace is running

The subscription list could be one of the items you can provide an answer on.

I agree that a node's subscriptions should be detectable with a Matrix perception test; they're connections to other nodes that involve significant data transfers (since as of SR4A you can do data requests to interact with other nodes without requiring a subscription), and that amount of traffic should be fairly apparent if you're specifically looking for it.

#2: There's actually a chart on page 222 of the 20th Aniversary book (not sure what page in the original 4e book) that gives sample device ratings, which lists normal vehicles as Rating 3.

Guns don't have computers in them unless they are a smart gun, or have a built in commlink (from arsenal.)
Smart links would be a good example of a periferial node, meaning it is very limited in what it can do and is not intended to run a lot of software. (I think the diffrence is described in SR4A and Unwired, but basicaly they have only admin level access, can only run special software intended for them, and only 1 program. They only start working like real nodes/commlinks if they are clustered. Clusters are in Unwired.) Most cyberware and electronics equipment are also periferial nodes and are intended to only be linked to a PAN on their user (which tends to be a Commlink as the primary setup.)
A gun with built in commlink is unsuprisingly just like a Commlink.
Drones and vehicles (even non-rigger adapted ones) have commlink like computer systems in them and are intended to run multiple programs and act somewhat autonomously.

For systems other then commlinks and nexi, the Device rating for the gear is assumed to be it's rating of Response, Signal, System, Firewall, and with drones/vehicles also it's Pilot.

You are referring to this? p. 148

"A standard weapon is considered a “dumb� device with no
Device rating and no wireless connection, so it cannot interact
with the Matrix nor can it be manipulated through it. Th is is
the case for most melee and thrown weapons. Many 2070-era
fi rearms, however, are equipped with a smartgun system, either
off -the-rack, as an internal or external fi rearm accessory, or as
a weapon modifi cation. Th ese weapons are considered smart
weapons with a Device rating of 3 and a wireless connection,
as well as the basic equipment that comes with a smart system
(laser range fi nder, small camera, sensors to keep track of heat
buildup and ammunition, an automatic gun mode switch, and
clip ejection). Some of the weapon modifi cations listed below
require a weapon to be a smart weapon."