Speeding Up Hacking, A Matrix rant Options Track this topic Email this topic Print this topic Download this topic Subscribe to this forum Display Modes Switch to: Outline Standard Switch to: Linear+

[Garrowolf]

I was just looking at cetiah's post of his hacking rules as a way of speeding up hacking and I thought I would say something about speeding up hacking.

I understand that the official position of Fanpro is that hacking occurs by proxy of personas because the matrix just moves that fast. I don't disagree with the thought that computers will move faster and faster but I don't like the consequences of the way they thought this out.
Okay if they are saying that the persona is the one actually doing things because we are too slow then why not do that all the way? Make hacking a function of agents and not people. Then you can throw away all the matrix rules and have an agent roll. The rest occurs behind the scenes.

> Well that wouldn't be fun and it would take away a character type. People who are interested in computers like to be able to deal with the matrix.

But the matrix system doesn't reflect actual computers. If you know much about computers then you know that most of this doesn't make any sense.

>But it could make sense in the future!

SO your making a detailed computer system to please the people who are interested in computers by making one that doesn't make sense to people who know computers? ( See Lewis Black for an expression about this logic, the one with the confounded jowl shaking)

I think that computers will be fast and are fast now but they still depend on people. They will be limited to what their computer is telling them. No one can stare at something at the rate these people are talking about. Making a perception test several times a second against all the traffic in a system for a normal 8 hour shift? Think about that for a second. You will end up will a drooling mess that can see anything anymore. It won't work. You are effectively increasing the time that they perceive by a factor of their initiative increase. So they are staring at something for effectively 24 hours! Even staring at the same area for more then a few minutes normal time will wear out a person. How is it going to feel with high resolution hyperactivity of the brain from all the input?

So we go back to programs making the perception tests. The spyder only pays attention when the programs tell him something. That way the poor spyder doesn't fall unconscious with a few minutes of his shift starting.

Now programs can make constant perception tests. Okay so for one why would the roll vary? They are artificial so they would notice the same amount of information all the time. So take it's average successes as a threshold.

So we have eliminated the problem of multiple perception tests by the system. If you are going to do something that they might think is strange then you just have to beat a perception threshold with your stealth.

We have sped things up alot so far! Now you don't have to worry about making perception tests for every second.

More in a bit.

[cetiah]

[Garrowolf]

So at this point we can reflect several actions in a system by one threshold. I think we are making progress! So let's see if we can continue this trend.

We can easily say that the programs are doing several actions at one time and this is reflected by one rating.

So now we have the idea that one program can do several things that can be reflected by one rating.

So we take these two ideas and put them together and say that one program interacting with another program can constitue multiple descrete actions to possibly counter each other. Therefore we could reduce the complexity of what is going on by saying that one program can negate another program through a series of actions that we as people could follow if we tried. Therefore we can further say that the interactions of the software is were the speed of the matrix occurs and we do not need a proxy. The give and take of the individual programs takes up that task.

Therfore we can go back to the idea that a person CAN interact and do things on the matrix if they have the right software and we DON'T need a proxy to do those actions for us. We just need to have a copy already loaded to do something.

So the matrix opens back up for actual people. These people can do actions that will either work immediately or fail immediately. We don't have to try and move as fast as a computer. We just need to think ahead.

So what does this mean?

We can slow down hackers and speed up hacking!

Hackers don't need as many IPs. They can do their actions as fewer rolls if some of the rolls are built in as threshold modifiers to a smaller number of rolls. This means spending a little more time with each roll but far fewer rolls. You roll to gain access to the signal, then you roll to do whatever action you need to do. You don't need to roll for the system hardly ever. No perception rolls, no firewall rolls, no agents wandering around rolls, and no exhausted spyder's rolls. You either do the action or you don't do the action. That is all the GM needs to figure out.

If you reduce the number of rolls needed then you can reduce the IPs for the hacker. This way they don't slow the game down for the rest of the players. Ideally you can have them act with a normal number of actions on their turn and that will be it! The hacker's action no longer becomes the best time to run to the grocery store.

[Garrowolf]

[cetiah]

Update: 1

First of all, let's be clear about this. I agree with you. The basic assumptions underlying the Matrix are dumb, especially if you can no longer put them in a "virtual reality" context. But I also understand why they are there and how they relate to genre. I'm a little fuzzy on the source material, but I'll try to explain how I see this:

[Serbitar]

Well as I elaboreted in some other threads, the matrix can be modeled to be exactly what you want it to be. There is always some fluff handwaving argument to justify everything. Its virtual reality. For example I could counter Garrowolfs argument about staring at traffic in that way:

You dont stare at the traffic, you know it. Using the sniffer (or whatever) utility it is directly transfered into your brain like a knowsoft and you just know what traffic is there, depending on the rating of the utility, so you are able to process it.

Again, you can justify everything if you want in VR. So the modelling is completely up to you. I can also justify whay the system/node doesnt know everything that is happening in it. It has no routines to do so. Thats what agents are for. Agents carry the intelligence. Of course you can again say: "Agents should know everything the node knows if authorised to do so". But again, things can be faked. And thats exactly what a hacker is doing. Disguising as something else.
To say it the third time: If you are creative you can justify everything. Just be consistent. So the actual game reality of the matrix has nothing to do with its mechanics.

Thats why the simpelest Matrix rules are those, that have a dice roll related to every desicion you make. Of cours you can ask why even roll, but then you are questioning the paradigm that in RPGs you roll to do something if this something is important and could fail and will ahve some consequences if it failed (if something has no consequnces, it should never be rolled for).

So in the matrix you roll dice if you want to do something that should, for any reasons, have a good (say at least 5%) to fail.

But you have 2 cases. One, wehere you are ding this action with access rights, and two, where you are doing this without. The second test has to be more difficult or we would be facing the paradoxon of some people with high hacking and low computer skill doing things better illegaly than legaly.

So we need two kinds of tests. One closed test, and one that is harder. Again tehre are two choices for this: Either make it an open test, where the ndoe rolls against you, or just subtract a number related to the node directly from the result.

Finally you only have to consider the actions which should be rolled fore. The more actions you consider "roll-worty" the more complex your system become, the more possibilities you get and the slower it will be.

Thats just about it on what to consider when making up matrix rules:

- you can justify everything, so you can focus on game mechanics
- to make game mechanics as simple as possible connect only desicions - and consequnces - to rolls
- make it consistent in itself
- make it balanced so that it is consistent with your game world

The rest is just a matter of personal taste: What is considered worthy rolling for.

[cetiah]

[Serbitar]

Main line of thought is more like:
Only roll for every desicion -> further reduce the number of rolls by sorting out those desicions that are important to you or your rules system

Definition of desicion: Something that somebody can choose to do or choose not to do, while both actions would have some benefit and some drawbacks over the other.

If a choice has only benefits or only drawbacks over another choice, there is no desicion. You just logically do what its clearly best.
Of course the boarders in this definition especially what is considered a drawback or a benefit are fluent.

[cetiah]

[Serbitar]

Well, perception test do not really fall into the whole "desicion" classification. Of course everybody wants to percieve everything and it is thus not a desicion to percieve (everybody will want to do it alwayswith mostly now drawbacks).

Its good to bring that up. Apparently there is another reason to roll dice for: Something that has a large impact on the outcome of a scene even if there is no desicion that triggers them.

So I would roll for perception if the outcome of the test is influencing the outcome of the scene. And it definitley influences things if you either notice the stealthed IC or not. (Or whether you can evade a trap or simmilar examples)
But I would agree with Garrowolf that one roll should be enough. Multiple rolles that cover the same thing can always be transformed into one test with a lower or higher threshold. Its like your "one roll per scene" rule. After all, you can transform probabilities after X rolls into the same probability after one roll, by manipulating dice numbers, or thresholds.

This can of course change, if the number of dice rolls is not known. For example in my rules, an IC that spotted you, rolls perception tests every time you do something illegaly. I can not transform this into one roll up front, because at this point it is not know how many illegal actions the hacker is going to take.

So the whole thing is not that easy.

[cetiah]

[Garrowolf]

Well one of the things that I tried to figure out when deciding my matrix rules was what was actually being seen. Let's break down the idea of a perception test for a moment.

You are testing to see if there is a user in the system. This would be a yes/no. If you only have one user then the chances of being noticed doing anything are significantly higher. But let us just think about a system that has a decent amount of traffic - say some corporate server.

Okay the system has a connection that is crossing the firewall so it knows that there is traffic from outside. If the hacker has supressed the firewall then they have gotten this far. The firewall is the primary determining factor on if a person belongs as far as I cn tell. The system has been spoofed to that point at least.

Now unless you have nested firewalls (which you theoretically could) then you are considered a normal user with basic access. The security at this point knows you are in the system but doesn't consider you strange.

We can handle this part with an access roll. You got in and can navigate around the system.

What can a matrix perception test on the part of the security system tell it at this point? Well you have gotten access and presumably not set off any alarms (glitched your roll), so it thinks that you belong. Will it watch you closely? Probably not if there is a normal amount of traffic in the system.

Okay so you decide to act strange but not do anything illegal. What does the security do? Nothing. Keep in mind that a person with the ability to open multiple windows at a time and exist in several systems at the same time, which is entirely normal, will act like a distracted idiot staring at the walls. So what constitues strange behavior? And how does a security agent know what is strange for a human? I don't see any psychology programs for agents as the norm. If they aren't doing anything illegal and they have some importance in the world but not necessarily in this system then harassing them because they are acting strange will just get the programmer fired and all the agents will be nicer and we go back to he didn't do anything illegal.

So where does that leave a perception test now? Well you can't use it usefully to determine if they belong, you are already know that they are there, and you can't notice strange behavior. So what is left? You can tell if someone is doing something that would trip an alarm. But.... why bother if it is going to trip an alarm?

Anything that would fool the system is going to fool the agents the same amount or you are going to have to greatly increase the complexity of the system. An agent is basically the same as a robot and they can be fairly stupid when it comes to human behavoir. You are basically left with a series of alarms that the hacker may or may not be able to bypass. There is nothing from the system's standpoint for matrix perception tests to really tell you.

Now if you think about it from a users standpoint you still have problems. The interface supplies you with information that it thinks you would find useful but it will not display hidden processes. Therefor in order to find out the information that you want you have to change what the system is telling you. Now how does looking harder have anythingto do with this. I can see where matrix perception might either automatically open hidden processes to you or be used as a way of noticing a detail if you are overwhelmed with icons but I think that this can be done better other ways. Change your access level so that the information is available to you. Use a normal perception check since it is visual information anyway.

So what is the point of matrix perception from the user's stand point?

Not much.

I'm not talking about preventing people from noticing things in the matrix. I'm just saying that there are limits to what you are going to be able to notice in the first place and any information that is available is put right out there for you to see. I mean that is the point of VR in the first place.

So why slow down play with rolls to either see things you can't miss or things that you can't see?

[Garrowolf]

Cetiah, I think that you would be better enlightened by watching better cyberpunk movies. The Matrix movies are fun and all but they fell for their own illusions pretty quickly. They told you that the matrix was not real in the first movie and then believed what it was telling you in the second and third. They are very poor examples of VR systems because the point of the matrix was to fool people all the time.

Better VR systems would be Ghost in the Shell: Stand Alone Complex or even Johnny Mnemonic.

A good way of understanding VR is to think of video game programming. You have to process everything that is going on with the action in the form of numbers. You determine if an object is trying to occupy the same space and stop it's motion in that direction - you limit the grid location of the character from having the same values as the stationary object. Then you determine things like damage effect and check to see if the character can keep moving at all. Then you check to see if there is a state change in the character that will limit it's action or change the sprite table that it is using to reflect damage. Then you determine what objects the player can see. You use a process called ray tracing to start from the view point of the camera and backtrack to each object to see if it is in the line of sight. Any part of an object visible will be rendered. Then there is a second process or ray tracing that detects lighting that would shine on the object and changes the color of the object to match the expected lighting of it.

This is just a crude break down but notice how much of the process was occuring behind the scenes. Now if you were going to hack it you would do something like change the processes for clipping so you can walk through walls. You change the process for determining damage or set you armor rating to higher then the most powerful weapon. You could cause the ray tracing to only return a wireframe so that you could see through the walls.

From the programmers and the hackers point of view this is how the worl looks to them. They would look at the code, not the pretty lights. They could change the world radically very easily if they are interacting with the VR interface at all. If you can change the processes of the world then why does the display mean much of anything to you?

A better way to think about VR is it's a chat room. You have VR for social interaction. You don't need it to read a file. You can make it into a book if you want to but don't mistake that choice as something forced on you by the system.

[cetiah]

[cetiah]

I'd just like to point out the irony of this situation. The three people who've made their own Matrix rules are the three least likely to agree on anything Matrix related, because, let's face it, we've already proved how stubborn we can be when it comes to doggedly upholding our interpretations of the Matrix. :)

[Serbitar]

@Garrowolf:

Of course a perception test is useless when there is nothing to percieve. A hacked in user doing things that are legal within his hacked rights can not be identified as a hacker (unless by looking at logs how he hacked in).

But when you are actually doing a hack action, this can be noticed via a perception test. Why shouldnt it? You are doing illegal things with the node. Anybody in the node monitoring node processes can see it if he is good enough to pierce through the precautions the hacker is taking to obfuscate his actions.

Concerning VR:
VR is there to provide you information in a way that is more efficent than by looking at a code. Of course VR is not actually there. It is just there to "guide your brain", but that can be a drastic advantage, helping you to process stuff much faster and better.

[kzt]

[Brahm]

[Garrowolf]

What I was saying about the making a test to see if someone is seeing you doing something illegal is that all you have to do is have an alarm on the sensitive area. This would be alot more effecient then having a roving agent that happens to notice you doing something. Basically what I'm saying is that unless the hacker does something to cover up what he is doing then he should automatically be "seen". If he does do something and it is good enough then he shouldn't be seen. It is basically an unnecessary test (and therefore an unnecessary roll). The system would always be looking at the places that are sensitive is my point. But if you succeeded in hiding it then why test twice to see if you do.

I'm actually trying to make it easier on the hacker. Some people have suggested that the sysop or the system be making a perception check for every action the hacker does or every IP the server has. I'm trying to say that the system doesn't need to and it makes it much harder on the hacker AND it slows down the game for the GM and the hacker to be making so many rolls. We can make it game faster and make it so that other people are not so annoyed by hackers by finding ways to reduce the rolls.

[Serbitar]

Well, I am only suggestion one perception roll per action, which is disguised by stealth (which every hacker will use). If it is not disguised by stealth, any IC would auto notice it.
Futhermore, the system itself resits the hacking attempt, but that is something compeltely different.

[bait]

There are also alot of assumptions about how computers and networking work, however people fail to realize the amount of abstraction there is between the various levels of interface and the real system that lies underneath it.

[Garrowolf]

I have been trying to find anything that says that VR isn't a GUI. As far as I can tell from the descriptions it is. It may be a full sensory user interface but that is just because of the simsense added on. It is still a glorified desktop. All of your tools are 3d icons (that is what the game calls them). Instead of clicking on your program you pull it out of a sack or something. You could see a pdf as a book if you wanted to or put a floating window in the air.

I'm sure that there is a VR programming language. I remember earlier versions of the game actually had several programming languages listed. You would use them like creating an object in the sims. Then you could use the actual programming language to create an object or behavior that you don't already have coded. This way you could create that Cthulhu-Eats-Your-Head hat you always wanted. You could also create a lot of hacking programs and make them look like something normal. I personally think that the Matrix is going to look alot like the Sims. (actually this description may help me with new players)

The actual programming languages would still be a collection of denser blocks of repeated code for the visual behavior and the actual action codes which could be simplier. It is basically like coding at three different levels all the time: VR, Visual, Actual Coding. At no point would you need to mess with the binary (which would be a bad idea anyway because it would require actual compiling by archetecture).

About Snow Crash: I love the book. It is still a bad decription of what a VR environment would be like because you can't have true negative space in VR. What I mean by that is that you have no one to one correspondance with the physical world. You would never have miles of empty space out in the VR world because there is no actual VR world. Your commlink may have a room or a house in it in VR. A Server could have a shopping mall in it. A huge server could have a whole planet or two in it. If there was a one to one corespondance like in Snow Crash then they would cancel each other out in space.

Snow Crash sort of has the right idea about interacting with other objects. They take it in a strange direction at the end though. In it he correctly points out that in order to interact with another avatar they would both have to agree to the attributes and parameters of a combat system - specifically that a sword does a set amount of damage to their avatars and their avatars can take a set amount. He does this by tricking them at the door. What he would actually be doing is providing them with an outer avatar that obeys the rules he created but is invisible until it has to write wound effects on the outer avatar and then uses that as a way to not only eject them from the club but ban them for a period of time. That is the best I can figure on what he was doing but I don't see how this logic would carry to the motorcycle fight at the end unless he had these commands as a series of viruses.... Maybe

I don't think that injecting a little reality into virtual reality would break down the basis for the cyberpunk genre. Very little of the stories I have read in cyberpunk actually even had any kind of cybercombat. They were more about the dystopia of the setting. I've seen several that had hacking and counter hacking. It's not integral to cyberpunk at all. It was just a part of Shadowrun early on based on a some writers who didn't understand computers. Even they didn't always get it as confused as Shadowrun eventually did.

Keep in mind that the Matrix movies are not a good analog to understanding the Matrix in Shadowrun. In shadowrun it is an advanced internet for communications. In the Matrix Movies it was a prison to confuse the prisoners. You could use VR for both but these systems have totally different reasons for working the way they do.

Actually in the original script for the Matrix the people were not being used for their power, which is a silly idea, they would have lost most of it in those lightning arcs and they would have reactors still. They were being used as a neural computer network that the machines were running on in our subconscious. They needed people to believe normally because it was distrupting their processors. This actually makes alot of sense. Especially sense the whole lack of solar power stuff was disproven by the third movie when they showed a hovercraft piercing the clouds. They could have circumvented that problem if they had to. Even if they didn't know how to create a nuclear reactor by then they could have gotten some humans to figure it out in the matrix and then build it themselves.

BTW Cetiah I don't want you to think that I am trying to bash you. I just enjoy debating. I work as a security guard all night and I try to find any intellectual stimulation I can. Theorizing on gaming topics and trying to write science fiction are pretty much it.

[cetiah]

[Synner]

Just to clarify, VR in Shadowrun has always been an interface method and not the program code or underlying structure of the Matrix. In fact, this was described in VRI and II, as well as Matrix. Simply put, everything in the Matrix is a representation of the underlying code construct and every such item is represented by an Icon. Each Icon (be it a persona, agent, node, file, etc) has a code component and a header which contained not only a file identifier but all the relevant simsense code (in the case of VR) necessary to interface with it. In the case of SR4 this header also contains simpler visual/other sensory cues for use with AR displays. Neither VR nor AR are intended to represent the underlying code.

[Serbitar]

I would toss in the idea, that on top of AR and VR header information every matrix entity could also have knowsoft header information. So you just look at it and know the "class" and such of the content.

Something along these lines:

You have a data package containing a cook book.

In AR you would see the image of a cook book standing around in your virtual shelf when you chose to display your virtual shelf in your room.

In VR you would see a book standing in a shelf in a virtual room, displaying title and stuff.

The additional knowsoft header (you need a datajack for this) would make it that you just "Know" that it is cookbook X with this and that kind of recipes by Author Y, by just looking at it.