In a recent thread this was interpreted as "An independent agent can only interact with a node when it was uploaded to it first."

Now the question: How can an independent agent ever hack into a node, when it has to be uploaded to the node first to interact with it?
Not as obvious: How can an independent agent ever log onto a node (legally), when it has to be uploaded to the node first to interact with it?

(Apart from that there is no reason why an agent shouldnt be able to access other nodes like any other program does, but has to upload itself to the node to do so.)


Conclusion:
This phrase obviously only means, that the agent acting independently has to run on a node, not on a persona and can access any nodes it wants, just like a normal persona from there. This node can also be a commlink.

Comments?

If you think that works, then please answer my questions.

Please answer the questions then.

So they can ask for login, even hack into the node and initiate a file transfer to the host (upload of the agent), but can not ask not for a piece of data (browsing) or open a file on the node?

Please answer the questions.

The agent has been uploaded on node A by its owner. It wants to access node B. It sends a log-on request to node B. If node B accepts, it "sends" itself to that node. If node B refuses (or asks for some kind of identification), it can use some hacking utilities to disguise himself as a legitimate traffic or to crack the identification.

You know it's the same for the hacker : how can he access a node to hack it if he doesn't even have a user account on it ? Or maybe there's some kind of "guest" account that only allow login actions...

Because an Agent is, basically, a computer virus. A virus can try to get onto my computer without ever getting installed, but it will have no effect until it is installed.

Do you think a Hacker can use his programs without gaining access to the node? No. That's not how it works. Think of the nodes as physical places that you need to "walk into." A passerby on the street can knock on my door, try the knob, or even smash the window... But they can't go through my file cabinent until they are inside. They can't flip the light switch. They can't raid the fridge.

Gaining access to the node is the same. You can do it the legal way (knock, knock) or you can hack your way in (smash the window). The difference between a hacker and an agent is that the hacker can bring his "super powers" (hacking programs and deck ratings) with him, but an agent is limited to the "power level" of the node.

It's not signal traffic, it's foot traffic, and the entire Matrix model is based on that idea.

Since they are supposed to act as Searchbots, that's the dumbest fucking thig I've ever heard.

It literally says that zero times in the BBB, and Agents wouldn't be able to do the very thing that they are supposed todo if that was the case.

-Frank

The ignorance of this post is collosal.

Acessing a node != running on a node

Mostly because SR Matrix is totally different from today's computer systems on that aspect: your persona has to be present in a node to affect it.

You can't Attack a hacker in a remote node from your commlink. You can't Edit a file in a remote node from your commlink. Your persona has to be there (even if your programs run on your commlinks).

It's true that nothing clearly states if the agent can "run" on a node and have its icon on another node (just like a hacker runs his program on his commlink and has its persona on another node) or if it can only be present on the node it's running.

I'd go for the second hypothesis, but it's a question that should be asked to the devs.

Granted. But you are "present" if you are accessing a node. Your programs are limited by the response of the node you are running on. Those are very different things.

The node I'm running on is my commlink. It is coordinating my programs and getting my processes running. A copy of my icon appears on a node I'm acessing, and I can use my programs against it, but that doesn't mean that my persona or any of my programs are running on that node.

That's bullshit. Your persona has to be there, but it doesn't have to be from there. You can very much attack a hacker in a node remote from your commlink.

If I'm running on Node A, I can access Node B. The programs that I'm running on Node A don't add to slowdown on Node B, and aren't limited in Rating by whatever Node B is all about. Now another Hacker can be running on Node C and also access Node B. Neither one of us are using up the program slots of B, but since we both have copies of our icon in Node B we can attack each other.

-Frank

As the rules for the SR4 matrix miss any description of 'connection', you can't even describe a phone call.

Read my inital post. I am not of this opinion. But others are. This Thread is to convince them otherwise.

No, only users have Personas.
Agents have Icons, though.

So sad, must be a Turing thing...

So if I analyze a Hacker Icon and an Agent Icon what do I see that's different? Behind each one will be a Response, System, Signal, Firewall, and program ratings?

there are the same attributes, and both a persona and a agent is represented by a icon online. but saying therefor that a agent = a persona or that a agent have a persona is potentially a long stretch of logic...

An successfull Matrix perception test indeed allows you to seperate Persona from Agent, and Agent from IC or Sprite, etc.

AFAIS, you can't hack a Persona directly, but have to trace it back to it's home Node.
An Agent, you could hack immediatly.

Really, it's quite open and shut. According to the actual written rules, an Agent can hack into other nodes and can access nodes other than the one it is running in. This is not really up for debate.

Whether this should be the case I suppose is, but presenting your house rules in this thread as if they were "interpretations" is misleading.

-Frank

I apologize. I removed the offensive posts.

I would agree with that interpretation based on the text you quoted.

This depends on your interpretation of uploading. My interpretation of RAW was that if an agent hacked into a node, the act of hacking uploaded the program. A successful hack means the virus (err... agent) was successfully uploaded. An unsuccessful hack means its just sitting helplessly on your comlink's node staring at the scary firewall.

This would probably mean that if the agent was defeated in cybercombat there that it's over and done with. Agent deleted. Not sure if this was ever in RAW or not, but its a natural consequence of the interpretation and isn't contradicted in RAW.

The act of logging on uploads the program. Again, if that node is shut down, you're agent is toast.

Disagree. If it worked this way, there would be no reason for an agent to ever move off your comlink onto another node. Unless you're willing to completely disregard the paragraph about moving from one node to another, this can't be correct. The text clearly indicates that it was meant to be a limitations on the agents' capabilities/efficiency.

I think this was likely the original intent.

Agreed on all points.

I don't think this is covered in RAW for or against, but it is reasonable to assume it can send requests to a node to conduct activities that the node itself was made to facilitate. So it can request a piece of information if that node was designed to provide that information to any requests for it, such as search engines and the like.

An agent can't browse around a node to see what's there without going to the node, but the node can communicate its contents to your agent if it wishes. Your agent can't really open a file on that node, but I suppose there might be some nodes that present any dumb clients that communicate with it to use some of its applications - so in that case, it's more like the node's system is opening the program for you.

The Agent can hack into a node, because RAW says it can. It can upload itself for similiar reasons. Ravor has already mentioned how this works.


---

Are there contradictions? Yes, but they're workable. I don't assume that just because its in RAW it has to make sense.

Do I agree that it should all work this way? Not neccessarily. As Frank advised, I am sticking with my interpretations of RAW for this post.

Exploit doesn't matter - Hacking is an exception that was specifically mentioned as being possible. Scan doesn't require you to be on another node. I could see Trace working either way. Even if it did work by "node-jumping" (as opposed to just requesting information from those nodes), your agent is simply moving from node to node. As it does its performance is potentially decreased, depending on what nodes it is accessing. A hacker can use this to his advantage. To use Browse to get standard publicaly-available information from search engines is no problem; to use Browse to look at all the files on a remote computer requires the Agent to physically be there.

I thought the whole point of hacking was to gain access.

Where is the agent hacking the node from? From another node? Strange, why can it hack in from another node (very complicated) but cant do a browse action from there (very easy)?

Same for loggin on.

Jackpot!

But there is a reason: You dont want to pay the response cost (also mentioned in the RAW paragraph).

Strange, a persona can do exactly that. Whys that? Mystic energy from the brain?

So if RAW said pistols could shoot 50 meters and assault rifles could shoot 10 meters you would just take it and move on?

this is shadowrun. That is entirely possible. Internet people are no longer just trolls on forums, but often in real life, too.

We don't have agents in real life to compare this to. You have basically decided what you want an agent to be, and the RAW disagrees with you. You may as well be being annoyed that lasers have a limit to their range in star wars in space, when you don't think they should.

Conceptually, I think the idea was that the firewall was supposed to be between the two nodes, so to speak. The firewall is the only barrier to access, and once it is taken down the agent can logon (upload itself) to do other stuff that requires access to the node.

I don't believe either of the above statements to be in the spirit of the rules. It is obvious that RAW never had any intention of giving users benefits for having two comlinks and it wasn't taken into account when writing the rules. It seems obvious to me that in reading the original paragraphs that you quoted a node was meant to go from node to node fairly frequently and this was meant to be a limitation. What you are describing is a complication, not a limitation. And I think you know it.

I don't know. Somebody already 'cried foul' when I tried to suggest a "better" way of handling things, so I'm limiting my statements to only what I interpret from the rules based on the information you quoted in starting this thread. I have been restricted from making value judgements on RAW or suggesting alternative methods.

This is the way RAW works based on my interpretations. I don't know why. I don't see any reason to make ANY distinctions between rules for personas and icons, and yet, RAW does. Personas just work slightly differently according to the relevent sections of RAW you quoted. I'm forbidden from saying anymore.

No, I would invent a very complex system of interconnected house rules that I feel better represent these issues and enhance gameplay and post them up for others to contribute/modify. I believe in an 'open source' attitude toward game design.

But I'm forbidden from speaking of these things on this thread. Shhhhh. Don't tell anyone.

Nothing against that, but why must it log on? That is the question. Why cant it just stay where it is and just continue to acces the node like it did while hacking it.

Maybe. But I am claiming that the rules are contradictory and should be changed if they mean what you think they mean.

No, you can just say: OK your right RAW is either flawed or interpreted wrong.

I'm sorry, but since none of us are 2070s matrix engineers, we can't very well tell you exactly how it works, any more than we can tell you exactly why they didn't have wireless everywhere in 2060.

by the way, neither of those are contradictions. They are part of the definition of the things you're talking about. A persona can log into other nodes from his comlink, and an agent can't, in the same way that a mage can summon fire spirits, but a shaman can't. that's part of the definition of what a persona and agent is. This may not fit the way you envision it working, but that's not the system's problem.

Only a guess:
Because the baseline assumption behind agents were that agents = IC. All the limitations that apply to IC apply to agents. And the designers didn't like the idea that a single IC could protect every node that its home node could communicate with. They wanted cybercombat to be a local affair.

The underlying topography of the VR Matrix was being used to determine how agents work, not real life networking. If the designers had started with "what kind of things could a hacker do with just a copy of linux?" the entire Wireless World Matrix chapter might have turned out quite a bit differently.

(I have more to say on this issue, because even my house rules work this way. But I want to think a little more on how I want to communicate my opinion, so I'll try to post later tonight.)

No, you have been saying that the two paragraphs that you quoted from the book mean something they do not. Or that they don't mean something they do. Either way, you are deliberately misinterpreting RAW and then accusing others of contradicting RAW when they try to correct you.

RAW is seriously messed up in quite a few ways. But that's okay, really. I think of RAW as a sort of "1st edition" of the new Matrix rules. They made a lot of good decisions and its a dramatic improvement, which is all I require from a new edition. They rode with new concepts and new ideas and I'm not surprised if the whole thing is fundementally flawed. They couldn't start from scratch even if they wanted to because there's a whole host of fans that would be upset by that. They needed to combine conflicting ideas and make it work the best they could.

I feel that my custom hacking rules work better than RAW, but it would have annoyed me if the SR4 hacking rules had turned out like my custom version did. My first reaction would have been, "What happened to my Matrix? I liked VR 2.0!!!" I imagine a lot of people would have had that reaction.

I say both.

You are interpreting RAW incorrectly, purposefully ignoring the basic underlying assumptions that RAW is designed to support because you disagree with those assumptions.

I'll also say that the basic assumptions underlying RAW are flawed, and a system that was built around different assumptions would work a lot better, be more realistic, and contribute to better game play. Then I would refer you to this website: Cetiah's Custom Hacking Rules

Because developers ruin all the good arguments...

Ultimately, tinkers don't ask questions like this because we want answers, nor do we necessarily seek to influence the game's development. We communicate to interact with other people, see what other tinkers have done, and contribute to bettering other people's games if we can. There's a reason we choose to communicate with forums and get opinions from people playing the game rather than shooting off emails to developers.

(Yes, it's because the developers have blocked us. j/k)

Several reasons:

- answers take some month most of the time
- most of the time the answers arent very satisfying (my subjective opinion)
- this thread is mostly about getting opinions from others
- this thread is also about pointing out a potential problem to SR players (it turns out that there are two very different interpretations)
- I use this thread as reference for the problem and to make sure I get every aspect of it (thats often the intention when I open threads, to find whether there are any arguments that survive my counter arguments, again maybe subjective)
- an answer from rob et al. most likely will not clarify/change RAW anyways

And finally I just didnt think about asking the devs. Maybe Im just not the dev asker type.

Edit: Ah well cetiah summed it up very well.

Please specify.

If the two paragraphs say that an agent must be running on the node to interact with it, then this is a contradiction with the rest of the rules and itself. I am not sayin that the rules say so. In fact, you do. RAW can not violate logic and consistency. If it does it is a problem.

This is OK, but they should have thought about the consequences. As it seems, they didnt.

I have no problem understanding this, and it seems non contradictory to me. Perhaps you could explain the contradiction. It says agents do specific things, and have specific limitations, and that one of those limitations is that unless they are latched on to a persona, their program has to be running in a node for them to act on that node. It seems very simple, but you seem to want the same rules that apply to agents to also apply to personas. I see no good reason why that has to be the case.