Blade
Feb 8 2007, 05:05 AM
I've already asked the question to Fanpro... I'll let you know when (if) I get the answer.
cetiah
Feb 8 2007, 05:16 AM
QUOTE (Blade) |
I've already asked the question to Fanpro... I'll let you know when (if) I get the answer. |
Can I ask what the question was?
Blade
Feb 8 2007, 05:23 AM
If an agent has to be loaded on a node to act on it.
(For example, can I load an agent on node A (rating 6) and have it logon on node B (rating 2) while still running on node A and running his programs from node A.)
kigmatzomat
Feb 8 2007, 01:37 PM
QUOTE (cetiah) |
QUOTE | The main difference between an Agent and a Hacker is that if you crash the Persona you crash the node but an agent can crash without impacting the node. That's a vital difference since IC are agents and you don't want a server to crash just because the IC died. |
Why do people keep saying this? It's not in RAW anywhere! You're node does not crash if the persona crashes! In fact, nodes can't even crash. They can be shut down, but they can't be crashed. Only programs and OSes can be crashed, and crashing the OS is one way to shutdown the node. Your node is not shutdown when your persona is crashed.
|
Because it is RAW.
QUOTE (SR p.211 Personas) |
Attacks made against your persona affect the device/OS, though Black IC programs affect the actual user directly.
|
Meaning that if the Persona crashes, so does the OS. Black IC that causes damage to the hacker's meat have no impact on the Persona or the OS.
kigmatzomat
Feb 8 2007, 01:42 PM
QUOTE (cetiah @ Feb 7 2007, 11:30 PM) |
I didn't say they HAD icons. The quote says they ARE icons. My very point is that they don't HAVE icons and can't project them. |
The word "icon" means "symbol."
QUOTE (Dictionary.com) |
i·con /ˈaɪkɒn/ Pronunciation Key - Show Spelled Pronunciation[ahy-kon] Pronunciation Key - Show IPA Pronunciation –noun 1. a picture, image, or other representation. 2. Eastern Church. a representation of some sacred personage, as Christ or a saint or angel, painted usually on a wood surface and venerated itself as sacred. 3. a sign or representation that stands for its object by virtue of a resemblance or analogy to it. 4. Computers. a picture or symbol that appears on a monitor and is used to represent a command, as a file drawer to represent filing. 5. Semiotics. a sign or representation that stands for its object by virtue of a resemblance or analogy to it. |
In SR4 there is a paragraph devoted to it.
QUOTE (SR4 p.211 Icons) |
Your persona’s icon graphically represents you in augmented reality (and especially in virtual reality, see p. 228), and in most forms of Matrix communications (email, messaging, phone calls, etc). Whether you bought your icon off the shelf or programmed your own, icons are easily customizable with a library of features and you can trick them out on the fl y with diff erent animated movements, color schemes, mutable design elements, and other digital skin eff ects. Icons take many shapes and forms, from animated characters and anthropomorphic creatures to more artsy or abstract designs like mobile waterfalls or swirling color patterns. All Matrix-capable devices have default icons loaded in case the user doesn’t have his own—usually simple blank-white anonymous anthroform shapes, oft en emblazoned with the device manufacturer’s stylized logo. Occasionally, programs you have loaded will add additional elements to your logo’s look, such as the glowing green force fi eld of an Armor program or the blurring eff ects of a Stealth program. Altering or swapping out your icon takes a Free Action. |
An icon is nothing more than a freaking graphic.
kigmatzomat
Feb 8 2007, 01:54 PM
QUOTE (kzt) |
QUOTE (kigmatzomat @ Feb 7 2007, 04:07 PM) | Tell me why anyone would let me run my agent on their hardware. Tell me why I would need my agent to run on their hardware. |
Why do you let people install spambots on your PC?
Oh, that's right, you didn't let them, but they did it anyway. That's what attack agents are. They are compromised systems that the attacker controls to some extent.
|
Exactly. I'm fine with the concept of attack agents breaking in illicitly. Those are hostile applications running without the admin's permission. A-ok by me.
I'm opposed to the notion that for a Joe WageSlave agent trying to find decent Mushu Pork in Peoria having to execute on every search engine, restaurant review site, and chat board across the midwest!
Kiyote
Feb 8 2007, 03:08 PM
QUOTE (kigmatzomat) |
QUOTE (kzt @ Feb 7 2007, 07:43 PM) | QUOTE (kigmatzomat @ Feb 7 2007, 04:07 PM) | Tell me why anyone would let me run my agent on their hardware. Tell me why I would need my agent to run on their hardware. |
Why do you let people install spambots on your PC?
Oh, that's right, you didn't let them, but they did it anyway. That's what attack agents are. They are compromised systems that the attacker controls to some extent.
|
Exactly. I'm fine with the concept of attack agents breaking in illicitly. Those are hostile applications running without the admin's permission. A-ok by me.
I'm opposed to the notion that for a Joe WageSlave agent trying to find decent Mushu Pork in Peoria having to execute on every search engine, restaurant review site, and chat board across the midwest!
|
If Joe Wageslave has decided that he needs his agent to check every search engine, restaurant review site, and chat board across the Midwest just to find Mushu Pork, then what is the difference to him between having the agent accessing the nodes vs hoping from node to node? Either way you are communicating with all of those nodes and gathering all that data.
kigmatzomat
Feb 8 2007, 03:43 PM
QUOTE (Kiyote) |
If Joe Wageslave has decided that he needs his agent to check every search engine, restaurant review site, and chat board across the Midwest just to find Mushu Pork, then what is the difference to him between having the agent accessing the nodes vs hoping from node to node? Either way you are communicating with all of those nodes and gathering all that data. |
There's no to Joe WageSlave. He just wants his mushu pork.
So under what logical, rational framework would anyone make it mandatory for agents to upload themselves to search engines jus to get results? There aren't any that I can think of nor are there any RAW that state they do. And if someone thinks there are, quote it and give me a page number.
Having agents wander the net conflicts with copy protection. What happens if it gets destroyed or hacked? How long until it lets you load another one into memory? If it were to crash while running on Google's node, how would the software know if it was terminated or still on walkabout? You can't tell me that Fuchi will let Joe unleash dozens of 1,000:nuyen:+ agents onto the net.
That says to me that the "normal" operating condition for an Agent is to only run on personal or corporate systems where the Agent has accepted user rights. That way if there is a crash, the Agent loader can confirm that the program was terminated.
The way for that to make sense corresponds with all the other supporting evidence that says Agents can access multiple nodes simultaneously from their home node without having to upload their code to the remote system(s).
ShadowDragon8685
Feb 8 2007, 04:41 PM
What makes sense to you dosen't nessessarily make sense to us, or to the game designers.
Deal. Or apply to FanPro, whichev.
Dashifen
Feb 8 2007, 04:41 PM
Alright, I'm not going to post it, but read everything under Agents and Using Agents from pages 227 and 228. That's my baseline to answer your challenge, kigmatzomat.
I feel that the key point is that they're trying to make an equivalence between drones and agents. Whether you feel that makes sense or not, they put it in the RAW: "Agents exist independently of the user in the Matrix, and are the equivalent of Matrix drones. They are capable of piloting themselves to a degree and can comprehend complex orders" (p. 227, SR4).
This equivalence, to me, implies that Agents move from node to node as they work. If I instruct a drone to follow that cab, then the drone will move through real reality to do so. Therefore, if I instruct this Agent to follow that datatrail, then the Agent will move through virtual reality to do so. I'm not saying that this has to make a lot of sense to you, but that's the way it makes sense to me.
QUOTE (kigmatzomat) |
What happens if it gets destroyed or hacked? How long until it lets you load another one into memory? If it were to crash while running on Google's node, how would the software know if it was terminated or still on walkabout? |
I'll answer your questions assuming that Agents travel from node to node, as you seem to be refuting.
- If the Agent gets destroyed then it is destroyed. Data it collected may be lost and it may prompt a hacker to try to follow any datatrail breadcrumbs it may have left behind back to the original source for the purposes of either (a) finding the "owner of the Agent or (b) if it is the owner performing the trace, fining out where the Agent met its virtual demise. As for getting hacked, I don't think Agents can be hacked because an Agent doesn't create a Node and only Nodes can be hacked. That being said, Agents do have Firewall ratings (equal to Pilot) but I feel that this rating is generally for the purpose of cybercombat and other matrix actions in which Firewall is rolled.
- If you've copied the original agent program on your commlink and you've loaded the first agent into a node separate from your persona and, thus, caused it to act independently, then you can load the copy of that Agent at your convenience into your persona and/or a node separate therefrom. Running a program or agent is defined as a Complex Action so this seems to take approximately three seconds.
- I'm not sure it would. If your drone is off following a cab and gets shot down by go-gangers, do you immediately know about it? No. You have to go out and try to figure out what happened. Much like #1 above, this could become a side plot in the story leading the hacker towards the node of the Bad Guy who killed the Agent.
- Why would Fuchi care?
I'm not sure what you meant by copy protection and I'm not following your conclusion that the "normal" operating condition for an Agent is to run on personal or corporate systems with user rights. Agents would also run on public systems with user rights. Hell, it could hack its way in to public or private systems to get security or administration rights if it were given instructions to do so and had the programs to assist it.
I don't feel that the RAW supports Agents having a home node therefore I don't think they support an Agent staying in such a node. In fact, p. 227 refers to the Response attribute of an Agent being equal to the Response attribute of the Node in which it is running. This is a clear ruling that Agents can move from node to node. Do they have to? I'm not sure the RAW explicitly states that they have to, but I also don't think it explicitly states that they shouldn't. Thus, it's up to you as the GM of the world in which you play the game to make that distinction.
It's my feeling that the RAW is trying to make Agents like Drones and to even make those entities congruent to Spirits and Sprites, though the latter two entities have one major difference: if the Spirit or Sprite is destroyed the Conjurer or Technomancer have a mental link with their entity (assuming the Technomancer hadn't been booted from the matrix and failed his Resonance check to reconnect to is sprites). Thus, all of these entities have to move around from place to place or from node to node to operate.
Serbitar
Feb 8 2007, 04:49 PM
QUOTE (cetiah @ Feb 8 2007, 05:03 AM) |
Because the only reason Frank and Serbitar are arguing for this position is that it creates their Agent Smith problem of another thread and they can turn around and declare how unreasonable and stupid RAW is because of what (they believe) it says. |
No, because it is the only non contradictory interpretation, the intuitive interpretation, the consistent interpretation and the streamlined interpretation.
I am not making things up just to blame RAW and have fun criticising people.
Dashifen
Feb 8 2007, 05:01 PM
Why is it the only non-contradictory (etc.) interpretation? I think cetiah, myself, kigmatzomat and others have all advocated non-contradictory (etc.) interpretations. We don't always advocate the same one, but I don't see how they're any better or worse than yours.
Kiyote
Feb 8 2007, 05:06 PM
QUOTE (Dashifen) |
Alright, I'm not going to post it, but read everything under Agents and Using Agents from pages 227 and 228. That's my baseline to answer your challenge, kigmatzomat.
I feel that the key point is that they're trying to make an equivalence between drones and agents. Whether you feel that makes sense or not, they put it in the RAW: "Agents exist independently of the user in the Matrix, and are the equivalent of Matrix drones. They are capable of piloting themselves to a degree and can comprehend complex orders" (p. 227, SR4).
This equivalence, to me, implies that Agents move from node to node as they work. If I instruct a drone to follow that cab, then the drone will move through real reality to do so. Therefore, if I instruct this Agent to follow that datatrail, then the Agent will move through virtual reality to do so. I'm not saying that this has to make a lot of sense to you, but that's the way it makes sense to me.
QUOTE (kigmatzomat) | What happens if it gets destroyed or hacked? How long until it lets you load another one into memory? If it were to crash while running on Google's node, how would the software know if it was terminated or still on walkabout? |
I'll answer your questions assuming that Agents travel from node to node, as you seem to be refuting. - If the Agent gets destroyed then it is destroyed. Data it collected may be lost and it may prompt a hacker to try to follow any datatrail breadcrumbs it may have left behind back to the original source for the purposes of either (a) finding the "owner of the Agent or (b) if it is the owner performing the trace, fining out where the Agent met its virtual demise. As for getting hacked, I don't think Agents can be hacked because an Agent doesn't create a Node and only Nodes can be hacked. That being said, Agents do have Firewall ratings (equal to Pilot) but I feel that this rating is generally for the purpose of cybercombat and other matrix actions in which Firewall is rolled.
- If you've copied the original agent program on your commlink and you've loaded the first agent into a node separate from your persona and, thus, caused it to act independently, then you can load the copy of that Agent at your convenience into your persona and/or a node separate therefrom. Running a program or agent is defined as a Complex Action so this seems to take approximately three seconds.
- I'm not sure it would. If your drone is off following a cab and gets shot down by go-gangers, do you immediately know about it? No. You have to go out and try to figure out what happened. Much like #1 above, this could become a side plot in the story leading the hacker towards the node of the Bad Guy who killed the Agent.
- Why would Fuchi care?
I'm not sure what you meant by copy protection and I'm not following your conclusion that the "normal" operating condition for an Agent is to run on personal or corporate systems with user rights. Agents would also run on public systems with user rights. Hell, it could hack its way in to public or private systems to get security or administration rights if it were given instructions to do so and had the programs to assist it. I don't feel that the RAW supports Agents having a home node therefore I don't think they support an Agent staying in such a node. In fact, p. 227 refers to the Response attribute of an Agent being equal to the Response attribute of the Node in which it is running. This is a clear ruling that Agents can move from node to node. Do they have to? I'm not sure the RAW explicitly states that they have to, but I also don't think it explicitly states that they shouldn't. Thus, it's up to you as the GM of the world in which you play the game to make that distinction. It's my feeling that the RAW is trying to make Agents like Drones and to even make those entities congruent to Spirits and Sprites, though the latter two entities have one major difference: if the Spirit or Sprite is destroyed the Conjurer or Technomancer have a mental link with their entity (assuming the Technomancer hadn't been booted from the matrix and failed his Resonance check to reconnect to is sprites). Thus, all of these entities have to move around from place to place or from node to node to operate. |
I agree with what Dashifen is saying here. I just have one little thing to add about the difference between an agent and sprites. If your Persona stays on the Matrix, your independent Agent is still a subscriber of your persona and counts against your subscription list. Since there is that link, you would be notified if your agent crashed for some reason. Of course, once you log off and jack out you break that link and are relying on your agent to send hit results to your Commcode.
Blade
Feb 8 2007, 05:10 PM
I'm failing to see how the interpretation that agents have to be loaded on the nodes to act on it is counter-intuitive (I've taken it for granted before you came up with that question (maybe I've got a poor intuition rating)), streamlined, consistent (with what ?) and not-contradictory.
I'm not saying that your interpretation isn't any of these, I'm just saying that, with available information, both interpretations can be made and besides one's personal intution, there's no "killer argument" that would discard either of these.
The only thing that's sure is that there's one official solution, but I think that the best way to get it is to wait for the official answer.
Dashifen
Feb 8 2007, 05:23 PM
QUOTE (Kiyote) |
I just have one little thing to add about the difference between an agent and sprites. If your Persona stays on the Matrix, your independent Agent is still a subscriber of your persona and counts against your subscription list. Since there is that link, you would be notified if your agent crashed for some reason. Of course, once you log off and jack out you break that link and are relying on your agent to send hit results to your Commcode. |
I can see that. I think you could instruct the Agent to drop its subscription to your persona if you wanted to though. The only benefit being that the Agent then couldn't be tracked back to you, but then you wouldn't be notified of any problems it's having. I can dance to it.
Cheops
Feb 8 2007, 05:33 PM
QUOTE (Dashifen) |
I don't feel that the RAW supports Agents having a home node therefore I don't think they support an Agent staying in such a node. In fact, p. 227 refers to the Response attribute of an Agent being equal to the Response attribute of the Node in which it is running. This is a clear ruling that Agents can move from node to node. Do they have to? I'm not sure the RAW explicitly states that they have to, but I also don't think it explicitly states that they shouldn't. Thus, it's up to you as the GM of the world in which you play the game to make that distinction.
It's my feeling that the RAW is trying to make Agents like Drones and to even make those entities congruent to Spirits and Sprites, though the latter two entities have one major difference: if the Spirit or Sprite is destroyed the Conjurer or Technomancer have a mental link with their entity (assuming the Technomancer hadn't been booted from the matrix and failed his Resonance check to reconnect to is sprites). Thus, all of these entities have to move around from place to place or from node to node to operate. |
The question, as I understand it, is not whether they can move from node to node, but do they move their entire "being" from one node to the next or do they just send an icon like a persona does.
eg. You can subscribe/operate in up to system x2 nodes at once. So can sprites and drones. Does this also mean that agents can subscribe to up to system x2 nodes?
If no, then there is no problem if the agent phsycially moves from node to node. If the agent can subscribe and operate on mulitple nodes then you run into some problems.
eg. I load my Agent Smith onto my commlink and tell it to do a data search. It subscribes to the first search engine. If the agent moves from node to node there are now 2 copies, 1 in the original node, and 1 in the search engine. The new one can now subscribe to Pilot x2 nodes and the original can subscribe to Pilot x2 -1 nodes. The Agent Smith problem.
The argument that I accept, and those you are arguing against, is that the agent operates just like a persona or a sprite. Agents can access multiple nodes, but they do not move from node to node. While they are operating in a node they use the response of that node as their response. However, the agent is still "physically" in their "home" node. This means there is only one copy but multiple icons up to Pilot x2.
QUOTE (Cheops) |
If no, then there is no problem if the agent phsycially moves from node to node. If the agent can subscribe and operate on mulitple nodes then you run into some problems.
eg. I load my Agent Smith onto my commlink and tell it to do a data search. It subscribes to the first search engine. If the agent moves from node to node there are now 2 copies, 1 in the original node, and 1 in the search engine. The new one can now subscribe to Pilot x2 nodes and the original can subscribe to Pilot x2 -1 nodes. The Agent Smith problem. |
you run into that problem regardless. all you have to do is order your agent to not delete itself from the node it's "leaving", when it moves from node to node. "moving" something, in terms of computers, means copying it from one location to another, and then deleting it from the old location. once the agent splits itself by moving, have either the old copy or the new copy subscribe to a device you control by proxy (a device subscribed to a device subscribed to your commlink, however many generations deep you want to go). any program that's smart enough to handle itself in cybercombat is smart enough to figure out what device it should subscribe itself to.
Pyritefoolsgold
Feb 8 2007, 05:43 PM
I have to say, the bit in the book about agents moving independently from node to node is less clear than the freaking second amendment.
But still, you have to remember that Agents are NOT executables. This is a brand new world of computing, and you really can't describe it using contemporary jargon.
Dashifen
Feb 8 2007, 05:44 PM
Indeed, that fundamental difference is probably not going to be resolved. I see agents moving their entire being from node to node. Otherwise, there's no point to the rule about their Response rating changing as they move and it invalidates the equivalence that the devs are trying to create between Agents and Drones.
Personas and Icons are different, as evidenced to me by the fact that a Persona has an Icon. Agents are Icons, as someone else pointed out above. I don't allow Agents to subscribe to Pilot x 2 nodes because that's something a Persona does and not an Icon. When a persona subscribes to a node, then their persona icon appears within it. But when an Icon moves from node A to node, that icon is removed from A and enters B.
As Kiyote said, I think that the persona could probably keep track of its agents if the GM allows that to occur so that the agent in Node B could communicate with the Persona in Node A, but other than that, I see no reason to allow Agents to subscribe as a Persona does.
Look at it this way: p. 211 states that a persona is "a combination of programs that you use, in conjunction with your device's OS, to represent yourself to other users and nodes in the Matrix." Since an Agent is a Program in and of itself, then it cannot be a combination of programs in conjunction with a device's OS because it doesn't have a device and, thus, cannot be a Persona.
Edit: I also notice that it doesn't say that Agents can subscribe to Pilot (System) x 2 nodes anywhere under Agents on p. 227 to 228. While it is a logical extension if you want Agents and Personas to act in the same way, the RAW doesn't seem to support that viewpoint.
Dashifen
Feb 8 2007, 05:49 PM
QUOTE (Pyritefoolsgold @ Feb 8 2007, 12:43 PM) |
This is a brand new world of computing, and you really can't describe it using contemporary jargon. |
QFT
I agree that in today's terms, "going" from one computer to another is a copy then delete operation. However, I see no reason that in 70 years, after two catastrophic global network crashes, technology wouldn't change in such a way that what we descibe today as a computer worm would one day be called an Agent.
Cheops
Feb 8 2007, 05:53 PM
QUOTE (Dashifen) |
Indeed, that fundamental difference is probably not going to be resolved. I see agents moving their entire being from node to node. Otherwise, there's no point to the rule about their Response rating changing as they move and it invalidates the equivalence that the devs are trying to create between Agents and Drones.
Personas and Icons are different, as evidenced to me by the fact that a Persona has an Icon. Agents are Icons, as someone else pointed out above. I don't allow Agents to subscribe to Pilot x 2 nodes because that's something a Persona does and not an Icon. When a persona subscribes to a node, then their persona icon appears within it. But when an Icon moves from node A to node, that icon is removed from A and enters B.
As Kiyote said, I think that the persona could probably keep track of its agents if the GM allows that to occur so that the agent in Node B could communicate with the Persona in Node A, but other than that, I see no reason to allow Agents to subscribe as a Persona does.
Look at it this way: p. 211 states that a persona is "a combination of programs that you use, in conjunction with your device's OS, to represent yourself to other users and nodes in the Matrix." Since an Agent is a Program in and of itself, then it cannot be a combination of programs in conjunction with a device's OS because it doesn't have a device and, thus, cannot be a Persona. |
Agents have Pilot = system. They have no other programs. Thus their response is determined by the node that they are operating on. In Windows terms, they'd have the response of whatever window was currently active. They don't need signal because they don't broadcast and they don't have firewall because they aren't a node.
Drones have all 4 persona attributes because they ARE different from an agent. Saying an agent is the same as a drone is wrong because you can't hack an agent like you do a drone. You spoof it like a wireless signal.
Alright, I can accept the argument that an Agent is an icon and it can't subscribe to multiple nodes. However, it still leaves the problem of "where" the agent is. Is it a cut and paste movement or a copy movement? You seem to assume it is the former.
What if someone programs an agent that doesn't cut and paste but copies? Presumeably that wouldn't be a hard thing to do. Regular programming task.
Self-aware agents like Knowbots and AIs would certainly find a way to circumvent the cut and paste.
"Do not call us Echo Mirage anymore. We are Legion."
FrankTrollman
Feb 8 2007, 06:17 PM
QUOTE (Dashifen) |
Why is it the only non-contradictory (etc.) interpretation? |
The interpretation where an Agent has to load itself onto a node to affect the node has a fundamental problem of interaction. That is, you have to affect a node in order to hack into the node. And you have to hack into a node in order to upload anythig into that node.
Now, hacking itself access into another node is an example of something that an Agent can do. It's in black and white on page 227, and in order to make the "Agents have to move to act" heresy function, its proponents have interpretted that as an
exception to the (unstated) rule that Agents can't act with the remote access to do anything other than upload themselves.
So... the statement that the Agent gets access by hacking another Node with Exploit is being pawned off as a double exception
- That the use of Exploit on a foreign node is an exception to the (unstated) rule that Agents can't affect nodes that they aren't on.
- That the Agent's gain of access to a foreign node is an exception to the stated rule that hacking and computer tests can be made on nodes that you have access to.
And sorry... that's just not reasonable. Cetiah is seriously asking me to believe that when it says that an Agent gets "access" it doesn't
mean that it has access, it only means that it opens up a file-transfer window that it can jump itself through but cannot send or receive any other information. Then he's asking me to believe that even though it doesn't ever say that an Agent can't use programs remotely and does give an example of an Agent using a program remotely that this example is actually a special exception that only applies to itself.
That's not a reasonable interpretation. At all.
-Frank
Cheops
Feb 8 2007, 06:26 PM
\signed
cetiah
Feb 8 2007, 06:35 PM
QUOTE |
And sorry... that's just not reasonable. Cetiah is seriously asking me to believe that when it says that an Agent gets "access" it doesn't mean that it has access, it only means that it opens up a file-transfer window that it can jump itself through but cannot send or receive any other information. Then he's asking me to believe that even though it doesn't ever say that an Agent can't use programs remotely and does give an example of an Agent using a program remotely that this example is actually a special exception that only applies to itself. |
Yes, exactly.
It doesn't matter what you think the word "access" means or what you think the word "icon" means. The point is that access and icon are defined as having a certain meaning in the system and when talking about agents your meaning of the words of 'access' and 'icon' don't apply, whether they are based on common sense, personal experience, or dictionary definitions.
To access a node is to move into it. Period.
You move through nodes in the Matrix. Agents do, persona icons do. Whether or not that makes sense to you, it's the way Matrix topography is defined in RAW.
You don't need to enter ("access") a node to hack into it, "hacking it" is the process of breaking into it so that you can enter ("access") the node.
What you're talking about isn't access by the book's definitions, it's connection. An Agent needs a connection to a node to hack into it. Once it hacks into it, it has access and moves onto that system. Thereafter, it adopts the stats of the node it is accessing.
Thain
Feb 8 2007, 06:59 PM
The rules are pretty clear that if you want an agent to operate in the Matrix indepenently, you must load it on a node seperate from your persona.
How is confusing in the least? I have read, and re-read, this entire thread... I also read, and re-read, "The Wireless World" chapter. I see no conflict between the written rules, the implied and explicit "fluff" and the way Agents behave.
You want an Agent to operate solo, you load it onto a seperate node. It may now operate on that node. Where's the issue?
cetiah
Feb 8 2007, 07:10 PM
QUOTE |
If you've copied the original agent program on your commlink and you've loaded the first agent into a node separate from your persona and, thus, caused it to act independently, then you can load the copy of that Agent at your convenience into your persona and/or a node separate therefrom. Running a program or agent is defined as a Complex Action so this seems to take approximately three seconds. |
I see nothing in RAW to support the idea that more that an agent can be activated more than once. In fact the reference to some goon's cyberlimbs possessing merely a single defensive program seems contrary to this view that the node could have 100 copies of that program loaded. Also, the reference that a user must load the agent into a node seperate from the Persona seems to reinforce the idea that the agent can only be loaded one place to a time. Also, the way IC is deployed. Programs can be copied but agents cannot. Thus, agents can't have agents loaded into them.
QUOTE |
I'm not sure it would. If your drone is off following a cab and gets shot down by go-gangers, do you immediately know about it? No. You have to go out and try to figure out what happened. Much like #1 above, this could become a side plot in the story leading the hacker towards the node of the Bad Guy who killed the Agent. |
I'm not sure if I agree completely with the reasoning here, but none of it seems wrong. I'm just not sure its the best analogy. Regardless, defeating an agent does not automatically trigger an alert to the user. There are rules (sort of) for that, too. In fact nothing short of crashing his OS or glitching on a hacking roll against him will 'automatically' trigger an alert to the user.
QUOTE |
I'm not sure what you meant by copy protection and I'm not following your conclusion that the "normal" operating condition for an Agent is to run on personal or corporate systems with user rights. Agents would also run on public systems with user rights. Hell, it could hack its way in to public or private systems to get security or administration rights if it were given instructions to do so and had the programs to assist it. |
I think this is a misconception regarding the agent's "files". It's connected the idea that moving through a node is the equivilent of copying its files, which isn't the case in Matrix topography. In fact part of the whole point is that the actual physical location of all files is largely unknown and irrelevent.
I may have contributed to this misconcerption early on when I said (incorrectly) that an agent was deleted when it was crashed while wandering off in a node somewhere. In actuality, there is no connection between where and how "files" are stored and what the program is or isn't capable of. Within RAW they are totally independant concepts. So the agent isn't deleted when crashed. It can be re-booted and loaded again in your persona or a node that your persona has access to (i.e., the node your persona is located in).
As far as I can tell, an agent can only be loaded once and must be either active or inactive. If its loaded in your persona, it's always active. If loaded on a node independantly of your persona, it can be stored as an inactive independant icon (waiting for some trigger event or signal to activate).
But the "files" used to load that agent are not actually transferred over. Since this provides no real drawbacks or advantages, the rules don't really cover it. But it specifically says that defeating an icon in cybercombat crashes it, and there are rules for rebooting crashed programs.
cetiah
Feb 8 2007, 07:16 PM
QUOTE (Thain @ Feb 8 2007, 01:59 PM) |
The rules are pretty clear that if you want an agent to operate in the Matrix indepenently, you must load it on a node seperate from your persona.
How is confusing in the least? I have read, and re-read, this entire thread... I also read, and re-read, "The Wireless World" chapter. I see no conflict between the written rules, the implied and explicit "fluff" and the way Agents behave.
You want an Agent to operate solo, you load it onto a seperate node. It may now operate on that node. Where's the issue? |
Two things:
1) Some people believe that an agent can project icons out to nodes in order to access them, regardless of where the agent is loaded. This is incorrect. They are making a distinction where none exists.
2) There has been some conflicts of interpretation regarding the meaning of "independant" and "the agent must be loaded into a node seperate from the persona". My interpretation is that it means "the agent must be seperated from the persona and loaded into a node" not "the agent must be loaded into a different node than the one the persona is located in". My own interpretation is also that "independantly" means "not loaded into a persona" not that its operating in a node other than your comlink's node, as was previously suggested.
Does that clear up the confusion at all?
Cheops
Feb 8 2007, 07:16 PM
The issue is what happens after you load the agent.
Is it allowed to move from the node that you loaded it to?
When it moves does it load itself onto the next node?
Does it then delete the copy that was loaded onto the previous node?
Or does the agent instead remain on the first node and send an icon into the second?
The confusion comes from the response of the agent = response of the node on which it is operating. This leads some people to the conclusion that the agent actually "moves" around the matrix, loading and deleting from nodes as it goes.
Others (myself included) seem to take the viewpoint that the agent stays put and merely sends out an icon as it moves around the matrix.
I think that this thread has gotten so entrenched that it is to the point where neither side is really going to budge much.
Thain
Feb 8 2007, 07:19 PM
Is it allowed to move from the node that you loaded it to?
Yes, it may move to any node it has access to or can hack access to.
When it moves does it load itself onto the next node?
It moves to the node, it is now limited to the levels of that node.
Does it then delete the copy that was loaded onto the previous node?
Or does the agent instead remain on the first node and send an icon into the second?
There is only one agent, when it changes nodes, the only thing left behind is a data trail.
I find that all to be pretty clear, I don't see the confusion...
cetiah
Feb 8 2007, 07:27 PM
Nice summary.
Let's see if I can clarify my position using this summary:
QUOTE |
The issue is what happens after you load the agent. |
The issue is also where the independnat agent can be loaded to. Can it be loaded in your comlink's node? A different node?
QUOTE |
Is it allowed to move from the node that you loaded it to? |
Yes, move but not copy. It can only access (move into) one node at a time.
QUOTE |
When it moves does it load itself onto the next node? |
Yes.
QUOTE |
Does it then delete the copy that was loaded onto the previous node? |
There is no copy in the previous node. There are no actual files being exchanged - the Matrix doesn't work that way regardless of your own knowledge of how programs are stored in computers today. The "essence of being" of the agent is loaded and copies of it cannot exist anywhere else.
QUOTE |
Or does the agent instead remain on the first node and send an icon into the second? |
No.
QUOTE |
The confusion comes from the response of the agent = response of the node on which it is operating. This leads some people to the conclusion that the agent actually "moves" around the matrix, loading and deleting from nodes as it goes. |
Yes, I have this conclusion. There's more evidence than just that, though.
QUOTE |
Others (myself included) seem to take the viewpoint that the agent stays put and merely sends out an icon as it moves around the matrix. |
Do you regard IC working this way? One IC program loaded anywhere on the system can essentially protect every connected node simultaneously? If its on a node connected to the Matrix, can it simultaneously protect every public access node connected to the Matrix?
cetiah
Feb 8 2007, 07:28 PM
Thain is too quick...
DireRadiant
Feb 8 2007, 07:30 PM
QUOTE (Cheops) |
Others (myself included) seem to take the viewpoint that the agent stays put and merely sends out an icon as it moves around the matrix. |
I think both cases can and do occur.
Agents can act independently, either connecting to other nodes, or moving entirely to other nodes and then connecting to other nodes.
kigmatzomat
Feb 8 2007, 07:33 PM
QUOTE (Dashifen) |
Indeed, that fundamental difference is probably not going to be resolved. I see agents moving their entire being from node to node. Otherwise, there's no point to the rule about their Response rating changing as they move and it invalidates the equivalence that the devs are trying to create between Agents and Drones. |
I believe Agents can move from node to node. I even gave an example earlier of when they should. I do not believe they must.
Actually, re-reading the section on loading an Agent, a strict interpretation doesn't actually says it can move on its own. It merely says that the owner can load it to a different node and that the Agent's rating is capped by whatever node you happen to load it onto.
DireRadiant
Feb 8 2007, 07:36 PM
QUOTE (kigmatzomat) |
Actually, re-reading the section on loading an Agent, a strict interpretation doesn't actually says it can move on its own. It merely says that the owner can load it to a different node and that the Agent's rating is capped by whatever node you happen to load it onto. |
So what you need is a pair of Agent boot strapping themselves all over the place to actually have them move.
FrankTrollman
Feb 8 2007, 07:47 PM
Cetiah, where do you find the RAW that states, or even implies, that "Access" means that you transfer all your source code to the node you are "accessing"?
Seriously, the basic opening statement on AR says:
QUOTE |
Augmented reality allows you to access data from the Matrix just about anywhere, overlaying it upon your physical senses like a personal heads-up display. |
I mean seriously. You access things from somewhere. You don't even more. You just... access the data from wherever you are. Why do you suppose that the word "access" suddenly has a different meaning for Agents? It doesn't say that the word means anything different for Agents than it does for people with meat bodies, so why would it?
Where's this piece of text that defines Agents as having a special case meaning of "Access" where for some reason it doesn't mean "Able to send and receive information and commands remotely" like it does for everyone else?
-Frank
Dashifen
Feb 8 2007, 07:51 PM
QUOTE (cetiah) |
QUOTE | If you've copied the original agent program on your commlink and you've loaded the first agent into a node separate from your persona and, thus, caused it to act independently, then you can load the copy of that Agent at your convenience into your persona and/or a node separate therefrom. Running a program or agent is defined as a Complex Action so this seems to take approximately three seconds. |
I see nothing in RAW to support the idea that more that an agent can be activated more than once. In fact the reference to some goon's cyberlimbs possessing merely a single defensive program seems contrary to this view that the node could have 100 copies of that program loaded. Also, the reference that a user must load the agent into a node seperate from the Persona seems to reinforce the idea that the agent can only be loaded one place to a time. Also, the way IC is deployed. Programs can be copied but agents cannot. Thus, agents can't have agents loaded into them. |
One agent can't, I had intended to mean that if you hack the copy protection (as described on p. 228) and make a number of individual Agent programs, you could load them up at your convenience. I was unclear, sorry.
Kiyote
Feb 8 2007, 08:13 PM
QUOTE (FrankTrollman) |
Cetiah, where do you find the RAW that states, or even implies, that "Access" means that you transfer all your source code to the node you are "accessing"?
Seriously, the basic opening statement on AR says:
QUOTE | Augmented reality allows you to access data from the Matrix just about anywhere, overlaying it upon your physical senses like a personal heads-up display. |
I mean seriously. You access things from somewhere. You don't even more. You just... access the data from wherever you are. Why do you suppose that the word "access" suddenly has a different meaning for Agents? It doesn't say that the word means anything different for Agents than it does for people with meat bodies, so why would it?
Where's this piece of text that defines Agents as having a special case meaning of "Access" where for some reason it doesn't mean "Able to send and receive information and commands remotely" like it does for everyone else?
-Frank
|
He would have gotten that definition of access from the books definition of access which is used with Personas and their icons throughout the book.
How about this section, titled Authorized Access:
QUOTE (SR4 pg. 215) |
AUTHORIZED ACCESS Every Matrix node has a set of authorized users, people who have the proper accounts and passcodes and are granted privileges to take certain actions on the node as legitimate users. Authorized users oft en don’t need to make tests when attempting certain tasks, whereas a hacker who infi ltrates the node would need to make tests in order to illicitly manipulate the node. Note that many nodes also have public access areas (or may be entirely public)—the Matrix equivalent of websites.
|
This section talks about the logging in of a persona to a node in order to get to the data on the node, which it refers to as accessing the node.
And just a little further down from there is:
QUOTE (SR4 pg. 215) |
Passkeys are one of the most secure ways of controlling access to a node.
|
Here we see the word access and connect used interchangeably.
QUOTE (SR4 pg.218) |
There is also a limit to how many nodes you can access at once: you can only connect to a maximum of System x 2 nodes at any one time.
|
Here we see that when you access a node, it means your persona’s icon is on that node. If you didn’t upload the icon to the node then how did it get there? If you are not on the node, then how can you be attacked in matrix combat?
QUOTE (SR4 pg. 218) |
Note that your icon appears in each node you access, and each “copy” icon may be attacked in Matrix combat.
|
All of these quotes form the manual use the word access, and that word is used t describe the process of trying to log onto the node, getting past the firewall, and uploading your icon onto the node. In VR this is represented by moving your VR persona icon from one VR theme to the other nodes VR theme. In AR this is done by merely changing what a window on your screen displays.
An agent has to go through the same steps. Try to log on, try to get past the Firewall. upload its icon. The agent is itself its icon, therefore it uploads itself. Thus it moves from one node to the other. If you were watching this in VR, you would see the agent move from one VR theme to the other nodes VR theme.
kigmatzomat
Feb 8 2007, 08:20 PM
QUOTE (DireRadiant) |
QUOTE (kigmatzomat @ Feb 8 2007, 02:33 PM) | Actually, re-reading the section on loading an Agent, a strict interpretation doesn't actually says it can move on its own. It merely says that the owner can load it to a different node and that the Agent's rating is capped by whatever node you happen to load it onto. |
So what you need is a pair of Agent boot strapping themselves all over the place to actually have them move.
|
Reminds me of a Dragonlance gnomish invention, the ratcheting ladder-lifter. You climb to the top of a double ladder, pull the lower ladder section up, which is held by the ratchets, climb to the top of it, pull the other section up, repeat until you loft into the sky.
Maybe that's what I'm missing. I tought SR4's matrix was sci-fi, with a basis in science. Apparently it's all minoi/tinker gnome technology.
FrankTrollman
Feb 8 2007, 08:24 PM
QUOTE |
Here we see the word access and connect used interchangeably. |
Uhhh... yeah. That's because they pretty much mean the same thing.
Since when does "connecting" to a node mean "you load all your sourcecode onto that node and delete yourself from the node you were connecting from, causing the connection to be broken"?
That's the disconnect, if you will, in this entire argument. We all agree that you connect to another node when you access it and then you may "manipulate the node." We all agree on that part.
But where does it say, or even imply, that your fucking source code has to get completely transferred to the target node before you can otherwise modify the data content of that node?
Why do you think that "accessing" allows and requires you to file transfer the entire agent but it doesn't allow you to file transfer 30 seconds of video feed? Seriously, what the hell is that?
-Frank
Cheops
Feb 8 2007, 08:31 PM
In the FAQ they clearly define the difference between connection and access.
To CONNECT to a device you must subscribe to it. This allows communication between the two nodes.
To ACCESS a device you must have the authority to do so. This is governed by the accounts.
So by definition Agents must be able to subscribe. They get loaded to a node. When they try to move (whether whole code or icon) they must subscribe to the node first and then try to hack it.
So the agent connects through subscription. At this point I think we can all agree that it is still on the home node and communicating with the new node. Agent doesn't have access rights so it must hack an account. Still on home node and assume basic priviliges are achieved. It is now communicating and has access to the new node.
To me it doesn't seem like the agent has to move to the new node to do so. It would still be on the home node but operating on the new one with response equal to response of the new node. It also implies to me that agents also follow the subscription limit of Pilot x2 rule that everything else does.
Which gets back to the question of how does it access multiple nodes at once if it must load itself onto each node?
As for IC. IC as defined in the book are a very special class of Agent. They remain inactive on the node until an alert is raised. At that point it loads and begins running its programmed response. This could include following a hacker through multiple nodes. Think of a nasty Black IC track you to several nodes and attacking you on all at once!
Patrolling IC IMO are basically just agents. They are always active on the node with orders to watch for unauthorized access. The designation as IC in this case is purely fluff.
cetiah
Feb 8 2007, 08:46 PM
QUOTE |
Since when does "connecting" to a node mean "you load all your sourcecode onto that node and delete yourself from the node you were connecting from, causing the connection to be broken"?
That's the disconnect, if you will, in this entire argument. We all agree that you connect to another node when you access it and then you may "manipulate the node." We all agree on that part.
But where does it say, or even imply, that your fucking source code has to get completely transferred to the target node before you can otherwise modify the data content of that node?
Why do you think that "accessing" allows and requires you to file transfer the entire agent but it doesn't allow you to file transfer 30 seconds of video feed? Seriously, what the hell is that? |
You are the only person whose even mentioned source code. No one else has brought it up or quoted it.
The very presence of source code anywhere in this example is your own addition and not derived from anything we've discussed or anything found in the rules.
kigmatzomat
Feb 8 2007, 08:51 PM
QUOTE (Kiyote) |
An agent has to go through the same steps. Try to log on, try to get past the Firewall. upload its icon. The agent is itself its icon, therefore it uploads itself. Thus it moves from one node to the other. If you were watching this in VR, you would see the agent move from one VR theme to the other nodes VR theme. |
For the love of.....
An icon is a graphical representation. Uploading an icon amounts to adding a .GIF image as your avatar here on DumpShock. Nothing in SR4 or conventional english gives "icon" any more meaning than that in relation to computers.
Here's an analogy. A real-world analogy that everyone should understand. It uses honest-to-ghost software that exists today.
The matrix is a game of Quake. It's a 3D environment you can interact with. you start Quake on your computer and begin in it's local environment. You choose your character imagery, modify the colors or even change the wireframe & animations. This is your Icon. The Quake.exe on your computer is your Persona. It's speed is based on your CPU (Response), OS (System) and security is based on your Firewall software.
You decide to go online and connect to a server. Some servers are private, requiring a password, some are public that just let you on. Once you connect, your computer tells the host what Icon you are using. If the icon is not in the host's library, your copy of Quake uploads it.
You then navigate the host's Vr environment, interacting as needed. Sometimes that means you download other user's custom Icons or just some data (like music, sound effects, video, etc).
Most players are users of equal class. Sometimes there are security ops who have the ability to kill your character, drop the connection, or apply a mod to the environment (a cheat). Sometimes there are admins, who can reconfigure the host's firewall to simply block any request you make.
Agents are AI. In quake this would be the bots, playing the role of IC. These are (limited) AI programs that do their darndest to kill you. They navigate within the VR environment natively and use weapons as best they can. They are typically run by the host.
In some cases, there are bot-servers. A bot server (http://www.gamers.org/dEngine/quake/Qbot/) is a separate AI running separately from the host or other users/players that can connect to a host. This bot does not tax the server except as another player. It doesn't have a full version of quake because it doesn't need to provide feedback to a user. it is on par with a hosted AI/Agents/bots on quality, capable of navigating through the VR and making decisions on tactics, but is not run on the game host, making the host perform better. If this is run by the game host organization, it is IC. If it is run by a player, it is an Agent.
Bot servers can be loaded onto any other computer and their performance is based on the computer running them. It does not include any user-interface features since it is AI driven rather than User Intelligence driven. The Bot can be given commands, in this case using script files, that change its preferences of weapons, movement, targets, and iconography. It can be given usernames and passwords to access remote systems.
Even though the user and the agent both access remote systems and upload their icons to the host, they do not ever upload their "operating" software (quake.exe and qbot.exe). There is no need to upload themselves to the host, as it would likely slow down both the host and the user/agent. The exception is where the host hardware is so fast that loading the agent on the host would improve the agent's performance.
Kiyote
Feb 8 2007, 08:52 PM
QUOTE (FrankTrollman) |
QUOTE | Here we see the word access and connect used interchangeably. |
Uhhh... yeah. That's because they pretty much mean the same thing.
Since when does "connecting" to a node mean "you load all your sourcecode onto that node and delete yourself from the node you were connecting from, causing the connection to be broken"?
That's the disconnect, if you will, in this entire argument. We all agree that you connect to another node when you access it and then you may "manipulate the node." We all agree on that part.
But where does it say, or even imply, that your fucking source code has to get completely transferred to the target node before you can otherwise modify the data content of that node?
Why do you think that "accessing" allows and requires you to file transfer the entire agent but it doesn't allow you to file transfer 30 seconds of video feed? Seriously, what the hell is that?
-Frank
|
FrankTrollman,
Allow me to clarify as it seems I have not adequately explained my views on this matter.
To start, allow me to make the following statement. Transferring an icon, does not mean transferring "the fucking source code". It is very clear from reading the manual that the source code of a program is a different entity from the running program. Pyritefoolsgold and other hit upon this earlier; an icon is not the source code it comes from and it is not an executable as we think of them. It is an icon. When you upload your persona's icon onto a node, you do not upload the source code of your entire persona, just the icon which is the persona's connection to that node. An agent's icon is the agent itself including the programs in its active memory and its orders.
As for Why I think ‘that “accessing" allows and requires you to file transfer the entire agent but it doesn't allow you to file transfer 30 seconds of video feed?’, that can be explain as follows. In order for that 30 second of video feed to be sent anywhere, a connection has to be made; the data needs to know its starting and ending point. In the Matrix, the connection is conversed in the login attempt and the firewall is the gatekeeper to making that connection. Once you have that connection you can then request the video feed. However there are no rules for the node to act as a server and send data to an icon on another node. What we do have are rules for an icon on that node to search for and download data from the node. In order for an icon to get onto a node it must be transferred to the node, this simple must be done as part of the connection or else the Matrix as we know it now, and knew it in SR3 doesn’t work. We are dealing with a VR which reacts in more closely to physical space then current data transfer standards.
Allow me to ask you this Mr. Trollman, if the agent icon is not on the node, then how can the node's IC scan for the agent? Magically an agent is allowed to access a node, uses a pass code, but since it has not icon, no physical presence on the node, it becomes immune to IC? IC can't scan for to detect his presence, cause he isn't there. IC can't attack him cause they have to go to the node he is on to get to him? that doesn't make any sense. Even persona's have to have their icon on a node to access data, otherwise IC would be useless.
--Kiyote--
cetiah
Feb 8 2007, 08:55 PM
QUOTE (FrankTrollman @ Feb 8 2007, 03:24 PM) |
QUOTE | Here we see the word access and connect used interchangeably. |
Uhhh... yeah. That's because they pretty much mean the same thing.
Since when does "connecting" to a node mean "you load all your sourcecode onto that node and delete yourself from the node you were connecting from, causing the connection to be broken"?
That's the disconnect, if you will, in this entire argument. We all agree that you connect to another node when you access it and then you may "manipulate the node." We all agree on that part.
But where does it say, or even imply, that your fucking source code has to get completely transferred to the target node before you can otherwise modify the data content of that node?
Why do you think that "accessing" allows and requires you to file transfer the entire agent but it doesn't allow you to file transfer 30 seconds of video feed? Seriously, what the hell is that?
-Frank
|
I really think the whole problem is that you have some idea of how accessing remote computers works and that conflicts with RAW about how one interacts with matrix topology.
Your own opinion is so ingrained into your perspective and so right that it seems intuitively obvious to you that the book must also be using these same sets of assumptions, regardless of what it actually says on the matter.
That is not, by definition, interpreting RAW.
That's using it as a basepoint and building up on it. Nothing wrong with that; but not what this discussion is about. This discussion is about how Shadowrun agents interact with the Shadowrun Matrix topology according to the rules and descriptions presented. It's unfair of you to be using your own ideas and presenting them as interpretation, bashing and insulting other people who don't know about your ideas and are interpreting RAW.
I strongly encourage you to seriously think of how much of your own knowledge and ideas of computers work are influencing your judgement in this matter.
kigmatzomat
Feb 8 2007, 08:55 PM
QUOTE (Kiyote) |
It is an icon. When you upload your persona's icon onto a node, you do not upload the source code of your entire persona, just the icon which is the persona's connection to that node. An agent's icon is the agent itself including the programs in its active memory and its orders.
|
Quote me the page in SR4 that says that agents "are" icons as compared to agents "have" icons. Quote me a page that says an icon is anything more than a graphic or other VR representation of an object.
Kiyote
Feb 8 2007, 08:56 PM
QUOTE (kigmatzomat @ Feb 8 2007, 03:51 PM) |
QUOTE (Kiyote @ Feb 8 2007, 03:13 PM) | An agent has to go through the same steps. Try to log on, try to get past the Firewall. upload its icon. The agent is itself its icon, therefore it uploads itself. Thus it moves from one node to the other. If you were watching this in VR, you would see the agent move from one VR theme to the other nodes VR theme. |
For the love of.....
An icon is a graphical representation. Uploading an icon amounts to adding a .GIF image as your avatar here on DumpShock. Nothing in SR4 or conventional english gives "icon" any more meaning than that in relation to computers.
Here's an analogy. A real-world analogy that everyone should understand. It uses honest-to-ghost software that exists today.
The matrix is a game of Quake. It's a 3D environment you can interact with. you start Quake on your computer and begin in it's local environment. You choose your character imagery, modify the colors or even change the wireframe & animations. This is your Icon. The Quake.exe on your computer is your Persona. It's speed is based on your CPU (Response), OS (System) and security is based on your Firewall software.
You decide to go online and connect to a server. Some servers are private, requiring a password, some are public that just let you on. Once you connect, your computer tells the host what Icon you are using. If the icon is not in the host's library, your copy of Quake uploads it.
You then navigate the host's Vr environment, interacting as needed. Sometimes that means you download other user's custom Icons or just some data (like music, sound effects, video, etc).
Most players are users of equal class. Sometimes there are security ops who have the ability to kill your character, drop the connection, or apply a mod to the environment (a cheat). Sometimes there are admins, who can reconfigure the host's firewall to simply block any request you make.
Agents are AI. In quake this would be the bots, playing the role of IC. These are (limited) AI programs that do their darndest to kill you. They navigate within the VR environment natively and use weapons as best they can. They are typically run by the host.
In some cases, there are bot-servers. A bot server (http://www.gamers.org/dEngine/quake/Qbot/) is a separate AI running separately from the host or other users/players that can connect to a host. This bot does not tax the server except as another player. It doesn't have a full version of quake because it doesn't need to provide feedback to a user. it is on par with a hosted AI/Agents/bots on quality, capable of navigating through the VR and making decisions on tactics, but is not run on the game host, making the host perform better. If this is run by the game host organization, it is IC. If it is run by a player, it is an Agent.
Bot servers can be loaded onto any other computer and their performance is based on the computer running them. It does not include any user-interface features since it is AI driven rather than User Intelligence driven. The Bot can be given commands, in this case using script files, that change its preferences of weapons, movement, targets, and iconography. It can be given usernames and passwords to access remote systems.
Even though the user and the agent both access remote systems and upload their icons to the host, they do not ever upload their "operating" software (quake.exe and qbot.exe). There is no need to upload themselves to the host, as it would likely slow down both the host and the user/agent. The exception is where the host hardware is so fast that loading the agent on the host would improve the agent's performance.
|
I find it hard to believe that this it the truth. After all it is the Icon that IC scan, it is the Icon that IC attack, it is the Icon that Black IC use to overload your simsense. It is obvious that the Icon acts as a connection to your Persona, and your persona acts as a connection to your system and meatbody.
Even in Quake, as far as the game is concerned, your avatar is more then just a picture. It has health, capabilities, and weapons. It is the conduit that other users and the game itself uses to get to you, the player. It is a connection.
Edit: Allow me to clarify, as I posted in haste. When I say obvious, I mean obvious to me. I make no such assumption that my views are universal or even correct.
cetiah
Feb 8 2007, 08:58 PM
QUOTE (kigmatzomat @ Feb 8 2007, 03:51 PM) |
An icon is a graphical representation. Uploading an icon amounts to adding a .GIF image as your avatar here on DumpShock. Nothing in SR4 or conventional english gives "icon" any more meaning than that in relation to computers.
|
That may be what you think an icon is based on your own real-world experience, but that's not RAW.
Would it have made better sense to you if they made up new terms that had nothing to do with your previous computer experience? "Cybercombat takes place between two jukfdlafghgos in the Matrix." instead of "two icons in the Matrix."
An agent is an icon.
That's in the rules.
How do you come to grips with the idea that an agent is a .GIF file? It doesn't work.
We're not discussing how icons should or should not work according to modern day computer standards. Icons have certain abilities and rules regarding them and in a discussion of RAW we should only be using those whenever possible.
Who really cares if your .GIF file or dumpshock avatar crashes?
cetiah
Feb 8 2007, 09:27 PM
QUOTE (kigmatzomat @ Feb 8 2007, 03:55 PM) |
QUOTE (Kiyote @ Feb 8 2007, 03:52 PM) | It is an icon. When you upload your persona's icon onto a node, you do not upload the source code of your entire persona, just the icon which is the persona's connection to that node. An agent's icon is the agent itself including the programs in its active memory and its orders.
|
Quote me the page in SR4 that says that agents "are" icons as compared to agents "have" icons. Quote me a page that says an icon is anything more than a graphic or other VR representation of an object.
|
- There is no reference anywhere that Matrix Agents have persona or are anything more than icons.
- All programs are icons. (I believe this is supported by RAW; I don't have the books with me. Someone else is going to have to verify this.)
- Programs loaded into your Persona change your persona's icon. (I don't have a page number on hand. It was quoted word for word in a previous post.)
- Page 226, "Hacking Programs", "Attack", "Armor", and the entire section of 230 and 231 "Cybercombat" describe combat between icons. (I believe this is deliberate. Agents and IC are icons. And personas have icons.)
- Page 227, "Using Agents". "Agents can [hack into nodes] (as independant agents." Here a deliberate distinction is made clarifying the matter. It never mentions they get a persona or follows rules other than those for agents, programs, and icons.)
- Reinforcing #5 with p.228, "Using Agents". "The [independant] agent will continue to operate [...] even if the Persona goes offline." (Here, it is clear that independant agents have nothing to do with the persona.)
- Reinforcing #5 again, pg.228 "Using Agents". "...you must load [the independant agent] on a particular node seperate from your persona." (Once again, no linking concepts with the persona whatsoever. A deliberate distinction is made. The agent does not share properties or characteristics of a persona.)
Why is it you feel that I need to prove my views/interpretation and you don't? Anytime you disagree with an interpretation and challenge it, you should provide an alternative reference to correct the situations. Because you could issue blank challenges all day long, regardless of how much text is quoted. All the above stuff refers to quotes that have already been posted.
kigmatzomat
Feb 8 2007, 09:30 PM
QUOTE (cetiah) |
That may be what you think an icon is based on your own real-world experience, but that's not RAW.
|
Oh really? Here is the RAW.
QUOTE (SR4 p.216 Matrix Jargon sidebar) |
Icon—The virtual representation of a program in the Matrix.
|
QUOTE (SR4 p.211 Icon) |
Your persona’s icon graphically represents you in augmented reality (and especially in virtual reality, see p. 228), and in most forms of Matrix communications (email, messaging, phone calls, etc). Whether you bought your icon off the shelf or programmed your own, icons are easily customizable with a library of features and you can trick them out on the fl y with diff erent animated movements, color schemes, mutable design elements, and other digital skin eff ects. Icons take many shapes and forms, from animated characters and anthropomorphic creatures to more artsy or abstract designs like mobile waterfalls or swirling color patterns. All Matrix-capable devices have default icons loaded in case the user doesn’t have his own—usually simple blank-white anonymous anthroform shapes, oft en emblazoned with the device manufacturer’s stylized logo. Occasionally, programs you have loaded will add additional elements to your logo’s look, such as the glowing green force fi eld of an Armor program or the blurring effects of a Stealth program. Altering or swapping out your icon takes a Free Action.
|
Tell me then, how according to RAW, that Icons are more than "graphically representations."
Pyritefoolsgold
Feb 8 2007, 09:33 PM
QUOTE (kigmatzomat) |
QUOTE (cetiah @ Feb 8 2007, 03:58 PM) | That may be what you think an icon is based on your own real-world experience, but that's not RAW.
|
Oh really? Here is the RAW.
QUOTE (SR4 p.216 Matrix Jargon sidebar) | Icon—The virtual representation of a program in the Matrix.
|
QUOTE (SR4 p.211 Icon) | Your persona’s icon graphically represents you in augmented reality (and especially in virtual reality, see p. 228), and in most forms of Matrix communications (email, messaging, phone calls, etc). Whether you bought your icon off the shelf or programmed your own, icons are easily customizable with a library of features and you can trick them out on the fl y with diff erent animated movements, color schemes, mutable design elements, and other digital skin eff ects. Icons take many shapes and forms, from animated characters and anthropomorphic creatures to more artsy or abstract designs like mobile waterfalls or swirling color patterns. All Matrix-capable devices have default icons loaded in case the user doesn’t have his own—usually simple blank-white anonymous anthroform shapes, oft en emblazoned with the device manufacturer’s stylized logo. Occasionally, programs you have loaded will add additional elements to your logo’s look, such as the glowing green force fi eld of an Armor program or the blurring effects of a Stealth program. Altering or swapping out your icon takes a Free Action.
|
Tell me then, how according to RAW, that Icons are more than "graphically representations."
|
The fact that IC can attack you through them?
You have to understand that the Matrix is all based on an early concept of cyberspace, one that is much more similar to actual space than the computer world we are used to. An agent moving from one node to another has very little to do with files and code and executables, and has more to do with the idea of a person just walking from room to room.