Help - Search - Members - Calendar
Full Version: The Matrix problems
Dumpshock Forums > Discussion > Shadowrun
Pages: 1, 2
sunnyside
ALright this is from Trollmans justification for his new rules.

QUOTE
Script Kiddy (where you can wave your credstick around instead of actually having any skill to hack effectively), Hackastack (where you can benefit from having multiple iterations of hardware to bypass structural limits of personal identity), Drop-Out (where you can choose to segregate yourself from the matrix and still hack effectively despite being unhackable in return), and Agent Smith (where you can gain extra actual actions from your pocket book).


Now some of these I know and some I don't. Searching seems to indicate that at least Hackastack wasn't thought up on these boards but it go brought in from elsewhere and isn't carefully defined.

At any rate could someone clearly lay out each of these. Bonus points for responding to the counter arguments before they're brought up.

For example I would take "Script Kiddy" to refer to having a hacker in a box. You simply have an agent do all the hacking for you. But this carries a number of disadvantages to (somewhat) mitigate this.
-no edge use
-no hotsim bonus, no specialization bonus. no quality bonus means that the agent will throw fewer dice than a really good mundane hacker and quite a few less than a hacking adept.
-if running out of a players comlink(such as during infiltration) one fewer program can be running at the same time.
-spoofable which could potentially lead to horrible things.
kzt
Script kiddy: You don't need any hacking skills to hack in SR4, it's all software/hardware ratings. So a bored 12 year-old with mom's credit card is just as effective a hacker as any 400+ point hacker.

Hackastack: When you your icon/hardware gets toasted you switch to another comlink and presto, a fresh, polished icon.

Drop-out: AR hacking. I black hammer the technomancer, he curses me and dies because he can't hurt me back

Agent Smith: I buy an agent, pop the copy protection off it and replicate it on a million toasters, which then hack into Ares corporate at the same time. One or more will get really lucky and succeed due to the law of truly large numbers.
sunnyside
QUOTE (kzt @ Aug 3 2008, 01:48 PM) *
Script kiddy: You don't need any hacking skills to hack in SR4, it's all software/hardware ratings. So a bored 12 year-old with mom's credit card is just as effective a hacker as any 400+ point hacker.


Well, except for the stuff I mentioned above. Yes? Meaning anybody can hack that way, but not as well. I kinda wish the difference was a bit larger though.

QUOTE
Hackastack: When you your icon/hardware gets toasted you switch to another comlink and presto, a fresh, polished icon.

Didn't they say hacking on the fly exploits weren't transferable that way? I mean at a basic level you can just reboot your own comlink to repair icon damage.

QUOTE
Drop-out: AR hacking. I black hammer the technomancer, he curses me and dies because he can't hurt me back


Well, good luck detecting the technomancer in the first place if they're competent. But yeah. With proper cyberware you're really only out the two dice.

QUOTE
Agent Smith: I buy an agent, pop the copy protection off it and replicate it on a million toasters, which then hack into Ares corporate at the same time. One or more will get really lucky and succeed due to the law of truly large numbers.


Did they ever differentiate between the processing power required to run something like Google and toaster? Or is that one still just left to the GM to enforce?

I've heard the agent smith thing before. And variants on it are certainly annoying. But one would expect the above to simply result in a denial of service attack as all ports shut down/system reboots.




Heath Robinson
There are really a few underlying problems.

QUOTE (kzt @ Aug 3 2008, 07:48 PM) *
Script kiddy: You don't need any hacking skills to hack in SR4, it's all software/hardware ratings. So a bored 12 year-old with mom's credit card is just as effective a hacker as any 400+ point hacker.

Problem: A clear case of Character investment failing to perform. The archetype shouldn't exist because the rules don't support the investments that the archetype makes having a good return. The archetype exists and clearly has a place in the fluff, so the system is failing to support genre emulation.

Solution: Change to the mechanics of computer usage to make wider usage of skills and attributes.

QUOTE (kzt @ Aug 3 2008, 07:48 PM) *
Hackastack: When you your icon/hardware gets toasted you switch to another comlink and presto, a fresh, polished icon.

Drop-out: AR hacking. I black hammer the technomancer, he curses me and dies because he can't hurt me back

Problem: These share the fact that recovery is cheap and methods to prevent cheap recovery do not feature, period. Even the Corrupt program is entirely vague on whether or not you can, in fact, permanently corrupt another user's programs. Should a GM actually do this, it will be perceived as being a bastard because it's not made clear in rules, nor in fluff, that the Matrix is not a happy funtime playground where Hacker and Spider engage in witty banter over the exchange of attacks, once the threat of Blackhammer has been removed.

Hack-a-stack is viable because individual losses do not matter, they cost nothing except time to recover from. You're best served by switching to a new link whilst the old one reboots into hidden mode. Drop-out is the same; I don't lose anything if the Technomancer crashes my 'link but the Techno loses everything when I Blackhammer him, and his mom for good measure.

Solution: Make program and OS destruction prominent tactics, make recovery costly but fast. Make it clear that Spiders are not called "Spiders" because they spin webs out of their arse, they're named after Spider Jerusalem and are as much of a bastard as he is.

QUOTE (kzt @ Aug 3 2008, 07:48 PM) *
Agent Smith: I buy an agent, pop the copy protection off it and replicate it on a million toasters, which then hack into Ares corporate at the same time. One or more will get really lucky and succeed due to the law of truly large numbers.

Problem: The assumption here is that the target HAS to stay on the Matrix, ignoring the general feature of Shadowrun that everything is symmetrical and they can do anything that you can. This is further rooted in the idea that an overt attack on a node can (and should) succeed in a professional security environment. The player also forgets that computer forensics (a Trace program, running off logs) would enable the corp to target the sources. Furthermore, the assumption is that no connection filtering happens.

Solution: A good chapter on practical network security. Point out to Players that most corps are willing to deny customers access to services and data in exchange for securing their vital secrets, blaming malicious hackers for the outage. Remind Players that the Corps still control the physical side of the nodes they're hacking and that Corps do not profit from caring what happens to their opponents (who have violated a list of laws as long as an OS source code). Allow firewalls to maintain blacklists and whitelists. Make offsite Spiders proactive in searching out the sources of incoming threats without logging into the Corp network. Give commlinks in the wild regular maintenance sweeps.


All these problems are fundamentally solvable so long as you stop making the Matrix quite so welcoming to Hackers. Again, I point to the fact that the Matrix was always fundamentally designed to allow Hackers to do their crap and that makes it all focused on letting Hackers do things that no sane Corp would let them. There is also the technological ineptitude of the originator of the Matrix concept (he said "That's stupid" when told how computers work) that holds back the introduction of sensible network security information for usage in the game.

"Therein lies a dog that will burn your brain out" is not good security when you can produce an infinite legion of disposables. That dog shouldn't ever get the chance to be let off its leash when a proper Hacker is involved, anyway. So making the dog redundant through proper access control is, fundamentally, denying the Hacker a chance to kill himself, and that will improve the game significantly.
sunnyside
Wait. In all of unwired they didn't bring back the ICE that frag your chips or send physical electrical bolts into your hardware?

Hmmm I suppose if the PCs got to use it that could be a problem.
sunnyside
Wait. In all of unwired they didn't bring back the ICE that frag your chips or send physical electrical bolts into your hardware?

Hmmm I suppose if the PCs got to use it that could be a problem.
Jaid
QUOTE (sunnyside @ Aug 3 2008, 06:01 PM) *
Wait. In all of unwired they didn't bring back the ICE that frag your chips or send physical electrical bolts into your hardware?

Hmmm I suppose if the PCs got to use it that could be a problem.

not really. there's a program that reduces commlink rating temporarily (ie until you reboot). and there's a technomancer ability to damage programs permanently. and i suppose you could potentially infect your opponent's software with a virus. but nothing quite like the old gray IC for hackers.
kzt
But since non-idiots keep off-line copies of their SW, this is just annoying. You just reload off-line.


To step back a bit, the fundamental idea underlying the matrix (as presented) is that everyone in the entire world lets random strangers execute unknown code on the their computers. In reality, getting your target to run code you provide is the "And I win" stage of hacking. (That's why you have antivirus software on your PC.) Hence it's an insane model.

Trying to fit the SR 2070 matrix in a 2008 book into a model created by a technophobe who refused to use word processors and instead used a mechanical typewriter for his first 3-4 books in kind of hard. The advances in computer technology and the way computers and the internet have totally permeated society makes most of the concepts kind of quaint. (The fact that everyone who even sees an SR4 book knows way more about how computers work than mr mechanical typewriter just adds insult to injury.)

But without abandoning the backstory to allow more rationality and common sense, I think Frank's current stuff is as good as it will get. You can always rewrite whatever you don't like.
sunnyside
QUOTE (kzt @ Aug 3 2008, 07:58 PM) *
Trying to fit the SR 2070 matrix in a 2008 book into a model created by a technophobe who refused to use word processors and instead used a mechanical typewriter for his first 3-4 books in kind of hard. The advances in computer technology and the way computers and the internet have totally permeated society makes most of the concepts kind of quaint. (The fact that everyone who even sees an SR4 book knows way more about how computers work than mr mechanical typewriter just adds insult to injury.)


That might have been an excuse back in SR1 or so but now it isn't. I think they keep the ICE and stuff around becuase it is more exciting and interesting than the RL model. And they at least dropped where it helps to be angry.

Especially since the RL model is really more like the drop-out/script kiddie side of things. Anybody in the shadow community could buy a new worm and then you execute in AR or a keyboard, whatever, and there isn't any way to damage hardware, the user, anything. And Agent Smith is pretty much exactly how denial of service attacks work.



kzt
There are some very clever people out there writing attacks that are very well targeted and very clever indeed. But, yes, it takes a while to do, they do it all remotely and don't post messages taking credit. They just want the money or the passwords, they don't want powerful people and organization knowing who did this.

The script kiddie stuff is just noise. We get tens of thousands of them every day The 2-3 clever attacks a week are what we worry about. That and the people going the "free p0rn" sites run by the Russian Business Network, who get some extra software installed along with their gay bestiality kiddie porn movies....
Heath Robinson
QUOTE (Jaid @ Aug 4 2008, 01:25 AM) *
not really. there's a program that reduces commlink rating temporarily (ie until you reboot). and there's a technomancer ability to damage programs permanently. and i suppose you could potentially infect your opponent's software with a virus. but nothing quite like the old gray IC for hackers.

The Corrupt program is very quiet about whether it can corrupt programs and what happens if you happen to be using such programs. Essentially, there are arguments in favour of it being able to totally trash programs permanently. The problem is that it doesn't come out and say either "yes" or "no", so any GM who actually uses it in play becomes a dick for blindsiding his hackers with something they did not think was going to turn up. This means that it doesn't happen, even if it should happen. The metagame concerns of the GM overrule the verisimilitude of the setting.

I mean, this would be alright if it was clearly, outright stated that Corrupt could not corrupt programs. My beef would then be that there was no oppurtunity cost for Hackers losing a commlink, hence hack-a-stack and drop-out are winning tactics because you cannot lose anything. The fact that the writer was so quiet about whether the Corrupt program can corrupt other programs (but there's the note at the end that corrupt can't corrupt the OS or compromise the integrity of the node that implies that it can) smacks of crunch time editing to remove the implication that Hackers can be deprived of their programs.


I would advise everyone who wants to comment on the Matrix to watch Dennou Coil, because the impact of getting hacked is serious in that series. Tell me, is the concern the kids display towards minor Matrix damage (well, it's equivalent in the setting) what you see your Hackers express in your games? Wouldn't you rather the Hacker be as concerned?
FrankTrollman
Script Kiddy

Your dicepool is as I'm sure you're aware, completely independent of your logic. The starting character is capped at getting +5 from their programs and +4 from their skills, and the ultra maxed character can manage to get a 6 or 7 in some skills and have a 6 in their programs. But a snot nosed corp kid or a character who isn't starting out but has a hundred grand to drop can get an agent and a program of rating 6. That is, they can take their online action telling their agent to do actions and rolling 12 dice, which compares quite favorably to anything any hacker could ever hope to do with their actual skills. While there are certainly some advantages to having any actual talent and personal capability, we seriously are talking about like 1 die over a low logic technophobe who happens to have thrown 20BP worth of cash at the problem. The disparity in resources it costs to go the "self" route versus the "equipment" route is quite stark, and the benefits are not. Which would actually still be OK if you were limited to one self and one set of resources, such that the skilled man with a pile of script kiddy programs was like twice as good as the script kiddy with script kiddy programs. But unfortunately...

Drop Out

"You" don't need to involve yourself with the Matrix at all. That is, in order to run the Script Kiddy programs, you can be in "AR" which means that the computer is running and you aren't connected to it. There is literally no unique output coming from your brain into your computer. It's just running "your" persona and you can choose to watch. Heck, you can choose to not watch. This means that "your" Persona can just run off and do its thing while you are in no danger at all. Hell, if you give it some macros and commands you can even walk away for a while, and come back to see if "your" persona has accomplished anything. And if it crashed in the meantime, it doesn't even matter because you can just reset it by pushing the reset button and moving on with your life.

Hackastack

You don't have to limit yourself to a single persona. Because you don't have to give any specific or unique input to run one persona, you can provide just as little to two personas. Or twenty-two personas. So long as you keep purchasing more commlinks, "you" can keep running new personas. You can send them all the same orders if you want, because you actually send orders with your gloves. What chooses to receive those commands is up to you. Unwired also allows you to stick commlink chips together in various exciting ways to completely eliminate sustained program caps. So with a large enough pile of commlinks (this pile still fits in a backpack, because commlinks are really small), you can not only run an unlimited number of simultaneous personas, you can run as many programs on each of your personas as you feel like running.

Agent Smith
The only meaningful restrictions on holding multiple agents limit the number of agents of the same set of source code. Now, over and above the fact that a Rating 6 Agent costs a crap tonne less than an actual hacker (see Script Kiddy above), there's the fact that even these restrictions have a simple and specific out: each Persona is allowed to run a copy of an Agent without worrying about the whole Access ID crap. But of course, as per hackastack above, "you" can have as many personas as you want!

So where does that leave us? It leaves us in a position where I can for less resources than it takes to actually be good at hacking myself, purchase a set of computers that will run two agent hackers that are individually about as good as I could ever hope to be.And any time I considered expending resources on personally being good at computers that would still be true. In essence, we are talking about a post-singularity world where it is the machines designed by our machines that actually do anything worth doing and our personal capabilities are completely meaningless.

Not necessarily an unlikely future. Just a future where having a hacking focused character is obviously pointless on first principles.

-Frank
sunnyside
QUOTE (Heath Robinson @ Aug 3 2008, 08:50 PM) *
The Corrupt program is very quiet about whether it can corrupt programs and what happens if you happen to be using such programs.


It's a rather moot point at any rate correct? They'll just load up a copy stored externally. At most I could see requiring a reboot.

@Frank.

Well that's the more elaborate writeup I was hoping for.

Now correct me if I'm wrong. But even a human at skill six can still get at least +2 for hot sim, +2 for a quality, and +2 for a specialization on the agent (and wasn't there some cyberware that helped?). And I'm correct about the edge and spoof comments right? That isn't something to sniff at from a PC point of view. A PC hacker is likely either at edge 5 or higher or is a hacking adept which gives a good +5 all the time. And spoof can potentially be insidious.

Of course I'll accept that isn't as stark an advantage as one would expect for having to run hot instead of in AR and investing a good 100+ BP. But at least there is that much.


However it seems like the hackastack and smith issues aren't so relevant in runs because of the central importance of stealth. More attacking programs means more chances the system will go on alert, in the case of a large swarm it may very well go on alert immediatly through an exploit glitch. Multiple breaches would translate to shutting down all connections or at least all wired connections. Now that's something, and still annoying for a GM, but it could have been done with a jammer anyway. And in the typical case where the hacker is trying to get paydata out of a secure system attacking with multiple agents is simply counter productive.

Interestingly what a swarm would be good for is rapidly aquiring and decrypting hidden nodes as the character moves along, which would allow hacking of hidden nodes in the same timescale as combat. I'm not sure if I consider that a good or bad thing.

sunnyside
tl;dr version. A real hacker can beat the secure system in the run and get the paydata. Agents can't do it.


Sorry to double post. But it occurs to me that the difference between a decent hacker and an agent is significant for practical numbers used.

For example take the classic datagrab. The team busts down the door and the hacker steps into the RF shielded R&D room at the heart of the compound.

This system is fairly spicey but not mil-spec so lets give it rating 6 stuff and a firewall of 9.

Now. Smithing is right out, and hackastacking is dubious. It gets to roll its system plus analyze (12 dice) against all trying to get in. So every additional attacker ups the odds it'll go on alert the first turn.

But even just a single agent. With a dice pool of 12 and stealth 6 the most likely outcome is that the agent is getting caught. As likely as not before it even gets inside for some automatic fail.

A good human hacker however could have (even at a day after start for the rating 6 stuff we assume above) skill 6+program 6 +2 for slinger quality, +2 for spec (this or stealth is what I'd spec in), and +2 for hot sim, then on top +6 for spending softmaxed edge. That's 24 dice with rerolls. Not a sure thing but odds are the system is breached on the first try, and then will fail to detect the intrusion.
tsuyoshikentsu
QUOTE (sunnyside @ Aug 3 2008, 08:04 PM) *
For example take the classic datagrab...


It doesn't work so well if you're trying to do it stealthily, yeah. But what you're actually doing is the RL equivalent of shooting it in the f***ing face instead of sneaking by it.

I mean, what are they gonna do to you, IC? You're in AR, remember? Heck, you don't even have to be in the room -- just leave the commlinks in there and let 'er rip. They trace, they run in, there are a bunch of cheap, anonymous, and disposable links in there -- and if it were me, a camera and a bunch of explosives.
sunnyside
What they're going to do is shut down all connections, totally shut down, or, since it doesn't even require a test or any time, delete a chunk of the most sensitive data. Note that while shutting down cleanly or cutting just a single connection out of many takes time or opposed rolls, running a command program to sever power to the antenna or the whole unit has neither problem.

At that point, worst case, the team is sitting in the middle of a corp compound and they blew it, they aren't going to get paid for this one. Though as the GM I'd probably have some kind of degree of success involved. Like they could still grab a prototype to get some payment.

Of course systems you can access from the comfort of your own home are a different story. But the corps know anything they put out there is easy to hack because of the "taking your time" hacking rules at the very least. The paydata is always on the other side of the RF paint at the least.

FrankTrollman
I see that the statistics are not strong with you.


First of all, the sample system you are throwing around is batshit crazy nuts and noone is going to be likely to get in. It rolls 12 dice looking for your Stealth in hits. And it's an extended test to activate that alarm that gets a number of shots equal to the number of rolls it takes you to get nine hits. What the heck is that supposed to prove? If you don't make it in with just one roll you probably set an alarm off if your Stealth is 8 or less. Getting in with one roll is something you expect iff you happen to be rolling with 27 dice (or to be fair: 16 dice and spending an Edge for a reroll). Even then you only have a basic account and the Firewall is still nine, so you're going to activate the alarm anyway when you try to do things that your account privileges don't allow. And of course, this could still fail spectacularly because the chances of the alarm catching while you're going in is substantially non-zero in any case.

So new plan: since you evidently are totally OK with a plan that sets off the alarm in this particular super-fortress that you are using for an example, let's just accept that and move on. Let's go for broke: we aren't accepting anything less than full Admin Privs. And we're sending a hundred Agent Smiths. We totally accept that we're setting off the alarm on the first round, and only a handful of them will actually get in there as Administrators on the second round. But we seriously don't even care at that point because with admin rights we take a single action to give full access to as many agents as we feel like sending and simultaneously lock out anyone else hoping to log in with real security access (thanks Unwired! Your poorly considered topology saves us again!).

Absurd? Oh heck yes. And actually quite pricey as well, since the price tag on each response chip required to get a new Access ID / Persona / Agent combination is no less than 8000¥, we're talking a lot of money. But the thing is that really, honestly every time you are confronted with the opportunity to pay 8000¥ towards giving yourself any matrix ability at all and paying 8 grand for another Agent Smith you'd have to be a god damned fool to choose any real ability.

But that's the point. Even with that piece of crazy edge case shenanigans, the matrix rules don't support quality over quantity. Sure rolling more dice causes you to average more hits; but any particular die roll doesn't roll average results. It generates some real number of hits that are above or below the average number. You are more likely to succeed with more chances and less dice than you are with more dice and just one shot. By a huge amount. And that's not even addressing the fact that straight up the rules don't even support having to hack into stuff as crazy awesome as the bad boy you were throwing around that almost made us consider hacking in with a brilliant maxed out character with tonnes of karma sunk into hot VR hacking. The Firewall cap of a Nexus only goes up to 6.

Quantity has a quality all its own. Statistically speaking, in Shadowrun 4th edition rules using the Unwired Overhaul, that quality is a much better return for expended resources than anything you could ever hope to ever do with personal badassery. That's just a mathematical fact. And hackers in Shadowrun, are supposed to be good enough at math to know that.

-Frank
sunnyside
QUOTE (FrankTrollman @ Aug 4 2008, 01:00 AM) *
So new plan: since you evidently are totally OK with a plan that sets off the alarm in this particular super-fortress that you are using for an example


Er. No I'm not. That's the point. Though honestly firewall 9 on a paranoid system is hotter than need be to express the point. Which is that with 50% more dice to start with and the ability to throw edge at the problem a live hacker can blow through systems on the first try far better than their script buddies (and again increasing the number of IC hacking in increases the odds that the system simply shuts down before they gain any kind of entry, or cuts power as one arrives in before its next action). And as system responses are left to the GM it seems rather reasonable that even slightly less paranoid systems would respond to multiple alerts going off with a full out attempt at going down NOW.


Now I'm not saying the RAW is awsome. Just that in RAW there is a significant and notable difference in what a live hacker can do over an agent or swarm of agents when stealth is the prime consideration.

Also, and I'm not sure about this so let me know what the deal is, if you were in a cybercombat situation, lets say a couple people are fighting over who gets to use the cameras inside a building neither having authorized access at this point, spoofing can be used by the live hacker or one of his minions to flip one of the agnets/comlinks to the live hackers side. Then turning it against the runner and his swarm. Of course now it's like two different agents doing the smith transformation thing to each other. But the fact the live hacker can't be spoofed gives him an asymetry that would give them an edge. Also they could generally be more trusting of their input, as who knows what the compromised agents are showing their supposed master.

Of course I suppose all that last bit is fairly moot. I'd never roll out a battle royal between two swarms like that anyway.
tsuyoshikentsu
QUOTE (sunnyside @ Aug 3 2008, 09:46 PM) *
Er. No I'm not. That's the point. Though honestly firewall 9 on a paranoid system is hotter than need be to express the point. Which is that with 50% more dice to start with and the ability to throw edge at the problem a live hacker can blow through systems on the first try far better than their script buddies (and again increasing the number of IC hacking in increases the odds that the system simply shuts down before they gain any kind of entry, or cuts power as one arrives in before its next action). And as system responses are left to the GM it seems rather reasonable that even slightly less paranoid systems would respond to multiple alerts going off with a full out attempt at going down NOW.

Now I'm not saying the RAW is awsome. Just that in RAW there is a significant and notable difference in what a live hacker can do over an agent or swarm of agents when stealth is the prime consideration.


Christ, do you actually READ people's posts?

I'd address the problems with this, but Frank already did. In the post above yours.

QUOTE
Also, and I'm not sure about this so let me know what the deal is, if you were in a cybercombat situation, lets say a couple people are fighting over who gets to use the cameras inside a building neither having authorized access at this point, spoofing can be used by the live hacker or one of his minions to flip one of the agnets/comlinks to the live hackers side. Then turning it against the runner and his swarm. Of course now it's like two different agents doing the smith transformation thing to each other. But the fact the live hacker can't be spoofed gives him an asymetry that would give them an edge. Also they could generally be more trusting of their input, as who knows what the compromised agents are showing their supposed master.

Of course I suppose all that last bit is fairly moot. I'd never roll out a battle royal between two swarms like that anyway.


Congrats! In the time it took for them to get one agent spoofed, at least two of the other successes have got in and locked everyone else out of the system.
sunnyside
QUOTE (tsuyoshikentsu @ Aug 4 2008, 01:28 AM) *
Christ, do you actually READ people's posts?

I'd address the problems with this, but Frank already did. In the post above yours.


Geez you people get confrontational about your matrix stuff. His point was presuming that a real hacker couldn't manage a system like that either. However even with the firewall at a hair raising nine an edge spending hacker simply can generate enough dice to give themselves a solid shot at getting in, finding the data, disarming the bomb and getting out. Especially if you don't need high clearances to access that data so there aren't many rolls.

With slightly less potent systems you get that the edge spender can fairly reliably hack the system without being detected while a hackastack is still probably going to blow out everything.

QUOTE
Congrats! In the time it took for them to get one agent spoofed, at least two of the other successes have got in and locked everyone else out of the system.


Alright, I don't have unwired, however I find it hard to believe they can lock everyone else out of the system. Maybe people who log in with actual codes. But in my example it was people fighting over nodes that aren't all that hard to hack in the first place so the spider can get in that way. If there is a way to lock all the spiders out how would the smiths be getting in anyway?

But again it would all suck at the table. Unless you really like epic icon fights and code up something to simulate it on your laptop.


The only RAW solution I see is using the variability of responses to greatly emphasize stealth, and extreme responses if multiple break in attempts are detected. Which, actually, isn't so unreasonable. It also would speed up play potentially as cybercombat can be slow.

I got nothing RAW that makes it so you wouldn't want to just operate in AR if you've got the reflexes for it though. Just the bonus dice. Though i suppose it isn't all bad if hotsim is only used for extrememly critical situations.

Blade
QUOTE (sunnyside @ Aug 4 2008, 09:05 AM) *
Geez you people get confrontational about your matrix stuff


That's why I won't post my point of view here: most Matrix haters here have had an intense hatred for the Matrix since the BBB came out and spent a lot of time complaining about it, highlighting the problems and interpreting the rules in the way that led to most troubles. In the meantime, most of the Matrix supporters have spent a lot of time trying to make sense of what was in the BBB and interpreting the rules in the way that led to the most consistent system.

So now it's really hard for any of them to be really objective and have a good constructive discussion about the Matrix, except maybe with those who were in the middle, but most of them were there because they didn't care much about the Matrix, so they probably didn't pay much attention to the rules and their potential problems.
Bull
QUOTE (tsuyoshikentsu @ Aug 4 2008, 01:28 AM) *
Christ, do you actually READ people's posts?

I'd address the problems with this, but Frank already did. In the post above yours.


This could very easily be taken as a personal attack. Play nice with others, please.
Ryu
The published rules focus a great deal on software. This is where they fail, the emphasize should be on skill and hardware. Using a linear cost structure (within the relevant bracket) does not exactly help, as benefit is often exponential (the effect of fixed TNs).

You determine Response, choose System, and then figure out what special combination of programs you want to run in a given situation. For high-end comlinks (which are cheap), you will never be lacking anything, as you now have software suites and ergonomic programs. If you know the rules, you can finally get away with using low-rated comlinks (and high-end software). But you don´t want to, because your user-proxy is not available with optimisation.

What does the focus on software net for the game? Attribute+Skill works fine for everything else. ACIFS for thresholds would not be half-bad and easy to get, anything that runs a persona would just need an MPCP-rating on top of that (yeah, old concepts). Agent Smith can be dealt with by limiting the number of hits autonomous programs may use, for example to (rating-2).
Heath Robinson
QUOTE (sunnyside @ Aug 4 2008, 04:41 AM) *
It's a rather moot point at any rate correct? They'll just load up a copy stored externally. At most I could see requiring a reboot.

Yes, load off external backups. Wait, the ones in the same network or room as the commlink that the Spider just fried a moment ago? Get outtah here, n00b!

Offsite backups are pricey contracts, but at least your programs are safe. Oh, sure, they have usage charges, but they're better than not having anything to work with. What was that? "But if they fry more than a few commlinks, I'm losing money?" No sympathy, omae. You choose to hack-a-stack, you chose the costs as well as the benefits.


And this works at least twice as well when OSes can be corrupted.
sunnyside
QUOTE (Heath Robinson @ Aug 4 2008, 06:21 AM) *
Yes, load off external backups. Wait, the ones in the same network or room as the commlink that the Spider just fried a moment ago? Get outtah here, n00b!

Offsite backups are pricey contracts, but at least your programs are safe. Oh, sure, they have usage charges, but they're better than not having anything to work with. What was that? "But if they fry more than a few commlinks, I'm losing money?" No sympathy, omae. You choose to hack-a-stack, you chose the costs as well as the benefits.


And this works at least twice as well when OSes can be corrupted.


Um. I'd be thinking along the lines of a chip you could just have on your belt or whatever. Offsite isn't doing you any good when you're on the wrong side of the RF paint.

And where are you getting fried comlinks from?

Heath Robinson
QUOTE (sunnyside @ Aug 4 2008, 01:19 PM) *
Um. I'd be thinking along the lines of a chip you could just have on your belt or whatever. Offsite isn't doing you any good when you're on the wrong side of the RF paint.

And where are you getting fried comlinks from?

All the more reason to not get fried when you're on the wrong side of the RF paint. A little risk makes for the spice of life, eh? The Spider could have mounted that chip you've got on your belt and corrupted all the data on it. Side effect of everything being wireless.

Fried commlinks were my idiosyncratic way of saying "compromised commlink". By that, I mean one which has been wiped clean of everything.
Ryu
Not everything is wireless. That paradigm can not work. Many devices can be turned off, for example.
Heath Robinson
QUOTE (Ryu @ Aug 4 2008, 01:44 PM) *
Not everything is wireless. That paradigm can not work. Many devices can be turned off, for example.

If your chip isn't wirelessly connected, then you have to spend a meat IP grabbing it and either toggling the wireless or inserting it into the commlink. Meanwhile, the target changes ID or calls in better security.

I support active computer security being scary as all hell. If you're not sneaking around as a user with the appropriate priveledges (afraid to so much as poke your head out), I would rather that you fail repeatedly when going up against any computer system larger than a doorlock. Once they boot you, you should not be able to use the same account. Accounts should take time and/or money to compromise (and, no, you shouldn't be able to make your own on the fly). I believe that supporting brute force attacks on all but the simplest devices is a grand failure of verisimilitude.

Once you are inside their network, they control your routing and they can just boot you messily by switching off the gateway you're using.
sunnyside
Something blade said in another thread got me thinking.

Each node in a hackastack costs a good 8K.

However in 4th edition 8K will also get you a fairly potent combat drone. But I don't think many people have had a problem with a full on smith like assault.

Why?

Well first of all stealth, but often enough players don't care about that. I think the primary issue is that they get blown up real good and you start losing money even if you complete the run.

However you stand to lose nothing with your hackastack.

So I wonder if a lack of Grey IC isn't actually a problem as opposed to just leaving out an old feature.

I figure the main issue would be player characters abusing it. But I think maybe the idea that some stuff only runs on ICE could be revisited, that worked pretty well back in the day I thought.
tsuyoshikentsu
QUOTE (sunnyside @ Aug 4 2008, 09:58 AM) *
However you stand to lose nothing with your hackastack.


Correct. Drop-Out is really the problem with all this -- because even an Agent Smith burly brawl could go wrong for a hacker if he got slammed with a black hammer.
Jaid
reintroducing proper gray IC is as easy as announcing that nuke has a version that lasts beyond rebooting the system.
Wiseman
The biggest thing about RPG's is that they require a gamemaster, and RAW just doesn't cut it against any decent rules lawyer or metagamer. All games have loop holes and exploits and its up to the GM to mitigate people breaking the game for their own self worship. When people quote RAW for selfish purpose (that is breaking all the imaginary supense, threat, and fun of all others at the table) then I point them to page 265.

That being said I'd like to note that I have a lot of respect for frank who is a an icon on these boards and has a swiftly capable mind to grasp some of the bigger loop holes in RAW. But I don't feel that any of these Matrix breaking effects survive unaltered at any serious GM's table. Of course my house rules and opinion aren't RAW, but that's really a moot point if you're sitting at my table to play.

A final note before the meat of the post. I don't have unwired yet so I cannot speak on what will and won't make it into the game. Also I spend more time playing than posting, though this forum has got me through many a long day at work.

Script Kiddies - So mama's boy bought an agent and yes the agent has some nice dice to roll. But the two largest draw back's to an agent is pseudo intelligence and the Spoof program. I'd like to focus on the pseudo intellect though: Agents are not perfect AI's that take commands and do whatever the player tells them to no matter what. Throw in a couple "does not compute" errors wasting an IP and they quickly lose their new shine while the player is making logic tests (albeit with 12 dice) against whatever threshold the GM is setting. Sure attack commands are basic, but what happens when the defending node throws something at the agent it just wasn't programmed to react to. Sure these are RP reasons but RAW on page 214 seems to give GM's quite a bit of leadway figuring out what is a "usual range of function" for the agent in question. In the end they simply do not have the imagination or creative problem solving abilities that hacking usually requires, and if the script kiddie doesn't know what he's doing, the agent is even less effective.

Drop out/AR hacking - I simply don't allow AR hacking to have more than one IP. Sure you can get wired reflexes and spend 1 of up to 4 IP's doing something in AR, but not anymore than that. As a programmer I can tell you that hands and brains don't always work smoothly and though i'm a pretty good typist, sometimes the faster I go the less gets done. Giving system commands usually requires exact language and mistakes amount to re-issuing the whole command string (or backspacing and such). I've heard the official ruling, but I simply can't let the Sam outhack the hacker and be safer too. This means a VR hacker will always be faster, hot or cold, and you're paying a stiff price for that safety.

Daisy chain of commlinks - You can't have more than one persona active, you're only one you. You don't get IP's for each persona because every commlink is hot sim or whatever. Even if you link them up for more storage, processing, backup comms, you're only actively linking with one comm and the rest are on standby (or using the connection of the one active comm since all the others are subscribed). If you crash the active commlink, sure you can just switch over to the next without rebooting but your connection is gone. Your persona appears in the next commlinks node and you have to hack all the way back in.

Agent Smith - I'm torn, some part of me thinks this is really cool, but the effect on the game just isn't tolerable. Since you can only have one persona active at a time then your limited in the number of agents you have. And considering agents aren't perfect themselves having an army of them flashing error messages could be worse than anything any other hacker could do to you. Also if your issuing commands to a group of agents (all attacking) it isn't that difficult to imagine another hacker spoofing a command to all of them as well (since the command seemed to come from you anyway), considering it's a complex action to load an agent back up, it could be an extreme deterrent to command them all to deactivate.

In the end I try to reward creativity but I also try to keep it fun and fair. It's a game of imagination after all so there really aren't any rules other than what is agreed upon when you sit down, and even that is open to some discussion.

To each their own.
ludomastro
Logical argument has a specific meaning. One that has little to do with arguing. That said, I feel that we can take a stab at it.

Given: RAW hacking has areas that are vague. (Please note that I do not have access to Unwired and therefore, can't speak to any rules therein.)

Given: Some players (and GMs) have issues with vague rules.

Given: Some players (and GMs) will automatically create interpretations to fill in the vague areas, some won't.

Given: The interpretations of the various groups will not agree, you have the seeds of conflict.

Question: How does the potential conflict get resolved?

Answer: One of two ways:
  1. Use RAW and make it work. (I believe that Aaron is arguing for this.)
  2. Use houserules instead.


Intermediate conclusion: They are essentially the same thing. Option 1 is just an agreed upon set of houserules while Option 2 is agreed upon houserules. Arguing about who is right is almost as bad as Technomancers shiving each other over whether or not it is Resonance or Disonance which is important.

My opinion: The only way to make any progress is to sit down and take a hard look at the RAW rules and determine if they are truly consistent. If not, then we either need to fix them or replace them.

Conclusion: Sorry, I can't provide a general conclusion because each of us has to look at the analysis to determine where we stand. Frank, however, made his statement, loud and clear. The devs made theirs as well. Time for you to make yours.

Mine?

I have never been a matrix fan: it was too dungeon crawl for me in earlier editions. However, I felt that the 4e matrix was going to work for me. Unfortunately, after spending some time reading the rules and noting that it became a sub-system with similar but not identical rules to the main game (I'm looking at you Program + Skill), I became worried. When I tried to make a Technomancer - because I like the parallels to mages and the fluff was cool - and failed ... due to the rules, I became disheartened. When I realized that the matrix was broken using the printed rules, I gave up.

I now handwave the matrix in all my games. (e.g. Hacker = NPC)

I don't own Unwired and have no plans to purchase it as everything I have heard about it seem to indicate that it is more of the same. However, I would like to have a set of rules that makes sense to me and that I consider playable.

Frank's rules give me that. However, I'm still kinda neutral on the matrix from a storytelling standpoint so don't expect my games to have a lot either way.

Anyway, this post has gotten far longer than I intended and I have to be up early tomorrow for work. Have a wonderful morning, afternoon, or evening - wherever you are in the world.
Blade
QUOTE (Alex @ Aug 6 2008, 08:07 AM) *
Answer: One of two ways:
  1. Use RAW and make it work. (I believe that Aaron is arguing for this.)
  2. Use houserules instead.


Most of the time, according to discussions on Dumpshock, such problems are actually treated this way:
  1. Interpret RAW in a way that makes it work fine.
  2. Interpret RAW in a way that makes it totally broken and then complain about it/make house rules.

And that can be the case even if the rules are clear and consistent, since, as with all communication, there is noise. There's a difference between the intention and the writing and between the writing and the interpretation of the reader.
So you end up with two sides: RAW supporters and RAW detractors. As usual when two sides argue about something, they convince themselves of the truthfulness of their point of view and further debates are getting nowhere.

Personally, I tend to try to get rules working instead of trying to understand them in a way that'd make them totally fucked up. That doesn't mean I don't use houserules. I use a lot of houserules, but only when my intent is different from the writer's.

For the Matrix, I agree that it was really hard to make sense out of the BBB rules and that they lacked some capital information to be playable without at least some personal interpretation. I don't know if that's because I already had done a lot of work on the Matrix, but for me Unwired added everything that was missing (and many new and interesting elements as well) and let you use the Matrix without the need for any house rule.

Sure it's still possible that some people will play differently with the same rules, but that's the case with every rules. There are groups out there where the GM will never apply modifiers that aren't listed on the tables and there are other group where the GM will add a lot of modifiers that aren't listed. Both follow the rules.
Ryu
The RAW can definitly be made to work. The extend of what I´d like to see is clear houserule territory, but a creative application of the RAW can make the unnecessary complexity of main-book RAW go away. Unwired offers quite a few options for doing just that (Open Source, Software Options, Logic + Skill, epic processor limits for clusters and nexi...).
ludomastro
QUOTE (Blade @ Aug 6 2008, 04:27 AM) *
Most of the time, according to discussions on Dumpshock, such problems are actually treated this way:
  1. Interpret RAW in a way that makes it work fine.
  2. Interpret RAW in a way that makes it totally broken and then complain about it/make house rules.


Agreed. However, I personally feel that having to interpret the RAW is bad for the game. It is my biggest peeve about the d20 rules for things such as grappling.

QUOTE
And that can be the case even if the rules are clear and consistent, since, as with all communication, there is noise. There's a difference between the intention and the writing and between the writing and the interpretation of the reader.
So you end up with two sides: RAW supporters and RAW detractors. As usual when two sides argue about something, they convince themselves of the truthfulness of their point of view and further debates are getting nowhere.


Communications theory made simple:
  1. What you thought.
  2. What you said/wrote.
  3. What they heard/read.


While I agree with you, that is why it is so important that someone outside of the creative process get called in to proofread and shoot holes in your creation. Makes for a stronger creation when you know 1 and can get 3'. (I say 3' becuase the process also works in reverse.)

QUOTE
Personally, I tend to try to get rules working instead of trying to understand them in a way that'd make them totally fucked up. That doesn't mean I don't use houserules. I use a lot of houserules, but only when my intent is different from the writer's.


I would love for someone to help me see the RAW hacking in a way that works. I don't think I would keep it anyway, because of where I want my game to go; however, I would still like to see it.

QUOTE
For the Matrix, I agree that it was really hard to make sense out of the BBB rules and that they lacked some capital information to be playable without at least some personal interpretation. I don't know if that's because I already had done a lot of work on the Matrix, but for me Unwired added everything that was missing (and many new and interesting elements as well) and let you use the Matrix without the need for any house rule.


Your statement is the first non-developer, non-CGL endorsement of the rules I've seen. If I were to get more feedback from other, I might give the rules a try. As it stand, I would rather spend the money on Arsenal or the new Companion.
sunnyside
QUOTE (Alex @ Aug 6 2008, 10:59 PM) *
I would love for someone to help me see the RAW hacking in a way that works. I don't think I would keep it anyway, because of where I want my game to go; however, I would still like to see it.


Like I said make stealth king. Systems that detect multiple intrusions figure they're humped and kill all(or at least all wireless) connections. Doing such(dropping all via a use of the command program to take out the antenna), doesn't take a test and couldn't be countered unless a stealthy intruder was already in.

Systems that detect a single intruder go a bit more paranoid than you probably have them now, especially ones with dirt paydata (runner figures they either get in clean or they're screwed).

The results. 1 Hackastack and smith are essentially no go for useful shadowrun stuff. As they'd inevitably set off alarms.

2. The larger dice pool and ability to use edge means the hacker has a significant edge over a script kiddie (and no spoofing). Since stealth matters they actually care.

Raw does leave you with the drop out problem. On the other hand they are two dice down on everything, and they have to get some solid cyber to be able to keep up.

Though the advantage of drop out(black IC immunity) is balanced out if you re-introduce grey IC. That's semi RAW, as it's stuff from previous editions. Sort of like how there aren't any rules now for military vehicles and ships. Doesn't mean there aren't any.

Synner667
QUOTE (sunnyside @ Aug 7 2008, 07:32 AM) *
Like I said make stealth king. Systems that detect multiple intrusions figure they're humped and kill all(or at least all wireless) connections. Doing such(dropping all via a use of the command program to take out the antenna), doesn't take a test and couldn't be countered unless a stealthy intruder was already in.

Systems that detect a single intruder go a bit more paranoid than you probably have them now, especially ones with dirt paydata (runner figures they either get in clean or they're screwed).

The results. 1 Hackastack and smith are essentially no go for useful shadowrun stuff. As they'd inevitably set off alarms.

2. The larger dice pool and ability to use edge means the hacker has a significant edge over a script kiddie (and no spoofing). Since stealth matters they actually care.

Raw does leave you with the drop out problem. On the other hand they are two dice down on everything, and they have to get some solid cyber to be able to keep up.

Though the advantage of drop out(black IC immunity) is balanced out if you re-introduce grey IC. That's semi RAW, as it's stuff from previous editions. Sort of like how there aren't any rules now for military vehicles and ships. Doesn't mean there aren't any.

That sounds more the Netrunning in fiction...
...Do as much as possible quietly, 'cos when the system comes after you, it'll be loud and dangerous.

And what skill is there is using brute force to get things done ??
The skill is in getting in, doing what you want, getting out - all without anyone knowing, until it's too late [bit like a well executed Shadowrun, actually].
sunnyside
Of course I suppose that still leaves some uses for piles of comlinks. Off the top of my head using them to continually do the electronic warefare thing to find and decrypt hidden nodes as the player moves around.

Though I'm not sure if that's a bad thing except for the dice issue. It means that if the player spends enough cash they can effectively hack during a run.

Though if a runner has a bunch of comlinks active and someone else just wanted to find one I'd really lower the threshold. And again grey IC play in here as runners would be nervous about losing piles of comlinks.
Ryu
QUOTE (Alex @ Aug 7 2008, 05:59 AM) *
I would love for someone to help me see the RAW hacking in a way that works. I don't think I would keep it anyway, because of where I want my game to go; however, I would still like to see it.

Your statement is the first non-developer, non-CGL endorsement of the rules I've seen. If I were to get more feedback from other, I might give the rules a try. As it stand, I would rather spend the money on Arsenal or the new Companion.


What are your specific problems with the RAW? I absolutely don´t like the main book matrix RAW, but Unwired is a totally different animal. (Don´t let yourself be deceived about that, not by someone who has a serious stake to loose if everybody thinks otherwise.)

All that is really needed IMO is
a) a limit on agents, Unwireds solution does not work => try giving agents a limit on hits
b) a simple explanation to your players that everyone needs a way to get free software, so it can! be glossed over.
c) the logic+skill capped by program rating optional rule.
d) a required personalisation of all skillsofts

Now we have done a few things on top of that, like removing the dumb extended tests that are all over the place... "try again" works similar for "your bracket of doable systems", while removing the unwanted auto-success.
tsuyoshikentsu
QUOTE (sunnyside @ Aug 6 2008, 10:32 PM) *
Like I said make stealth king. Systems that detect multiple intrusions figure they're humped and kill all(or at least all wireless) connections. Doing such(dropping all via a use of the command program to take out the antenna), doesn't take a test and couldn't be countered unless a stealthy intruder was already in.

Systems that detect a single intruder go a bit more paranoid than you probably have them now, especially ones with dirt paydata (runner figures they either get in clean or they're screwed).


THis makes hackastack BETTER. Under those rules, I can set two agents to completely DoS any given mainframe CONSTANTLY.
Blade
Why do you say that Unwired's solution for agents doesn't work? (Once you consider that the text about changing the AccessID will be errated out, according to what was said on these boards)
sunnyside
QUOTE (tsuyoshikentsu @ Aug 7 2008, 09:04 AM) *
THis makes hackastack BETTER. Under those rules, I can set two agents to completely DoS any given mainframe CONSTANTLY.


Newsflash, a hacker can knock out mainframes anyway. Just take a couple hours to probe, get in undetected, crash it or do whatever. Why aren't they knocked down all the time? Presumably people have better things to do than piss of Gridsec for no profit whatsoever.



Ryu
Cracked agents are dirt-cheap. I very much approve of cheap or free software (open source and all that), but it removes money as a balance mechanism. Agent strategies get "broken" in the single-digit range, so you could buy eight different ones.

Botnets work as advertised, and are balanced by the inherent danger for the user. Broad-scale attack on mostly random nodes, linked to your home node? MCT Matrix Security wants to have a word.
Cthulhudreams
Getting unlimited agents is merely a function of time. You just need a drone, and some autosofts.
Ryu
And a GM that is stupid enough to permit a Software autosoft for drones, if I get your method at all. "See, I even bought it mechanical arms so it can type on the virtual keyboard...."
Blade
Actually you can probably get the same result without autosofts or with monkeys instead of drones...
Given infinite time, they will produce an infinite number of agents, probably after finishing Shakespeare's complete works.
Heath Robinson
QUOTE (Ryu @ Aug 7 2008, 02:34 PM) *
I very much approve of cheap or free software (open source and all that), but it removes money as a balance mechanism.

I disagree for economic reasons. There are at least 10 big software houses in 2070, and they all produce functionally identical programs (this is canon). The price wars will have driven their program costs down to the bare minimum possible to support development.

The tricks 'n' traps deployed by corps against crackers will have only improved, and all the advanced statistical crunching and modelling software will make tracing the spread of cracked software back to the source much easier (helps when you actively go out and create botnets to watch the world). The risks of selling cracked software are high, the value of cracked software is higher than normal software (replicable, after all); the price of cracked (or, at least, replicable) software is, therefore, higher than normal software.

Then again, the business model implied by the "Registered" option is untenable and destroys the Software market. If you have a piece of software that will always be up to date without ongoing costs, what need do you have to replace it with a competitor's? The population in 2070 is relatively stable and everyone who is going to buy software already has (mostly, there's a little influx and outflow in the market). The software market is minute if you support free updates, and all the corps will recognise that. They have been around for a long enough time that they will be wise in the ways of the business.

The only way to profit in a world of free updates is product churn, whilst subsequently abandonning support for old products to incentivise switching to new programs. The alternative to is to charge for update access, which I think will be common in 2070. The first method is not supported by RAW (wherein you always get access to a better copy of the program for free if you have Registered, it doesn't run out ever), the second is definitely not support by RAW. By RAW you don't have a software market for anything but bespoke scuplting and commercial-scale number crunching.


I would suggest the following pricing modifications. Software costs its price to purchase outright, you don't get the Registered option but you do get Copy Protection. You can license the software for 10/20% of its base cost in immediate costs and a further 5/10% (needs tuning) every month. Licensed software comes with Registered and Copy Protection. Finally, you can get cracked (or free, as in "freedom") software, which has neither, for 150/200% of the normal cost.
Bull
This is why it helps to have easy to get along with players who are interested in working with the GM to get a good game and a good story, rather than simply abusing the rules. A good smack to the noggin with one of SR's nice, hefty hardbacks helps. smile.gif
sunnyside
It's functionally a house fluff rule. But the way I figure it is that, like viruses, it's sort of an arms race with code. In the sense that if they see enough of a certain type of attack they'll have patched it up in the next update.

Therefore to avoid heuristics most true hacker code is custom made. Unique like a work of art. Give out throusands of copies though and the next day it just won't work against a lot of corps.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Dumpshock Forums © 2001-2012