That's workable. Thanks. sounds like you could live with the idea then...

OK, here's the thing: I'm not talking about the feel, it's about seeing an opportunity and taking advantage of it. The Matrix of the 6th World is more akin to a modern electronic warfare battlefield than most modern net security; It's being able to see a weakness and exploit it before the chance slips away. Based on the descriptions of the attributes, that's more intuition: recognizing something without having to extensively consider it. But that's something I doubt we'll come to agree on regardless, so I'll leave it at that.


Yes, hardware matters, and I am not suggesting eliminating it in any way. I am just suggesting changing the focus. Hardware is still the BIG limit beyond the dice rolls themselves: maximum System which inherently limits programs in power and number and sets how tough the icon is going to be. Firewall is essentially a "hardware" aspect. Response is reaction speed and processor horsepower. That stuff is CORE in CyberCombat, so I think it's already central enough. I want skill and talent to have some place, and currently talent has NO place and skill has a limited place compared to everything else, especially without caps.

Because AccessID reduces the role of a human hacker and replaces it with automated defenses that just pull the plug when you get spotted, removing high tension scenarios where you try and defuse then copy the paydata off before the node resets - instead it just instantly axes you. Also: This is ignoring the rules exploit where you can just make anything impossible to hack - you have to observe the access ID in use before you spoof it to get in in the first place, so if people just log on directly there is nothing outside the node for you to observe so you can spoof. Which is dumb. But the first point: it removes tension and replaces it with automated plug pulling still stands which reduces the differential between a spider and an IC/Agent.

A mobile IC or pilot has another name: It's 'Agent'.

The reason to ban agents is that they just replace people. You can just set your agent to 'hack' mode and it will breach the pentagon and probably won't set off any alerts. When you can buy that, why have people? Agents just replace persona's with automatons and that goes explicitly against your design goal. Why do you even want agents anyway given that your design goal is 'more people' and agents just replace people. That's the explicit goal of an agent!

Kick the stupid things to the curb, they don't need to exist to make the basic rules work and they go against everything you want.

6 is just rules exploiting, it should be obvious from the explanation above of how rendering a node works. If it isn't, don't worry you've already house ruled it out by proxy.

Yep. It looks good. If you're using custom rules, making them smooth saves headaches on everyone's part. Anything else I need to crunch numbers on tonight?

That is a very good point. I could never come to terms with Agents having the Hacking and Cybercombat skills, and why doesn't it have Electronic Warfare? I could see a personal electronic assistant that can do some computing and data search being sold off the shelves, but why would it also be programed with the skills to hack and crash other systems?

Wouldn't that make them Linkers?

That said, my idea way back when was to just treat hacking almost exactly like you treat magic. Attribute + Skill for the actual tests with your commlink (not the programs themselves) determining the base "Force" of your actions. Programs would function just like spells; no rating themselves, but influenced by your hardware and your ability to use them properly instead. And it would emphasize the need for good hardware, good software and good skills. Agents would work in a similar fashion to Spirits, at least as far as how you use them and what activities they can perform to assist you. Your commlink would still limit how many and how powerful each Agent could be, similar to a Magic Attribute. Hell, that alone would fix a ton of the problems in the game regarding those stupid things. Couple that with "spirit powers" customized for the Matrix and include rules akin to Watcher spirits for common tasks, and you'd be set for the most part.

Program options would be tweaked a bit, but overall they would let you exceed some of your commlink's limitations at the cost of potentially putting yourself at risk via a Drain-like mechanic that crashes your commlink's stats directly rather than just your Matrix Condition Monitor, or potentially even your own meat body as a danger for "overclocking" your 'link. The flavor rationale for that being similar to using Hot Sim.

The really nice part is that the system would work just fine with Technomancers, too. The only difference is that they'd have rules for their version of initiation, conjuring and acting as a commlink. As opposed to the convoluted mess we currently have. To this day I still can't make a functional Technomancer without sitting there and reading over each rule a dozen times.. and still fuck it up in the end. It's ridiculous. Especially since, in reality, they've just come back full circle to the ideas of "naked decking" from 1st Edition. And, in the end, the difference between Technomancers and Hackers would be similar to the difference between a Street Samurai and a physical Adept. Same basic rules while actually playing, just slightly different ways and slightly different options for getting there.

Honestly I don't know why they didn't do something like that in the first place. I guess because so many of the developers think themselves tech junkies, they felt they had to continue to overcomplicate those rules for "realism." Which is pretty absurd compared to how other rules in the game work. But this is me starting to rant so I'll cut this part short.

Anyway, there really isn't a simple solution to the problem. You'll have to pretty much overhaul the entire system to make such a drastic change (which is what incorporating Attributes would be regardless of the method you choose) work properly. If you do it half-ass, you wind up with all kinds of broken and illogical inconsistencies across the board. Especially with all the bizarre twists and turns the rules take as they currently stand. But by overhauling it from the ground up and keeping it simple, you can avoid a lot of that at the cost of leaving players potentially confused and unsure of what to do as far as stats and gear go. That's a problem with any house rule, though. But nothing a little explaining and experience can't solve.

It actually makes a lot of sense. Breaking into a computer system has nothing to do with how fast you can hack at your keyboard or how good you are at elementary arithmetics, most of the time it just means waiting for an extremely complicated program to finish...or just call somebody and tell him it's the company helpdesk (IMG:style_emoticons/default/wink.gif)

So yes, I like the system because the formula "program + knowledge how to use it" gets at least partially away from the Hollywood idea that a pale nerd hacking like mad on his keyboard is all it takes to get into the super-secret government files.

I like this, as it gives TM another reason to shine:
Sprites.

Spites, are effectively agents, and in a world without agents, having access to sprites is worth a considerable amount.

If you want to keep agents, you'll want to severely penalize their abilities: limited number of programs, low agent rating, programs limited by agent rating, etc. Thus a complex automaton is expensive and difficult to build, and not nearly as effective as doing it yourself, BUT has the advantage of not being traced back to its user if caught.

As I recall, the stealth program is never a test though. THe stealth rating is the number of successes needed in a matrix test to detect you. This is one area that I have issues with.

Kenshara-my experience with the attribute+ skill (limit program rating) alternative in the matrix is that it has its benefits, but has its downsides.

Advantages: COnsistency with the resto f the rules set. There is also a balance factor that helps the GM better evaluate difficulty. Because it is consistent with the rest of the rules it makes it quicker and easier to run.

Disadvantages: The hard caps leads hackers to be screwed by low program ratings-so if it is not a 5 or a 6, why even bother with the program.
The Technomancer has the opposite porblem-it makes it impossible (b/c of threading) to make a system difficult for them. It is either a breeze (they can get rating 8 or 9 programs) or impossible. Note that this then has the downside of gimping a hacker (making it next to impossible for them) if you have a hacker and a TM trying to work together and you even attempt to make iot challenging to the Technomancer.


Note I do also use the optional rule that TM's CFs are like spells in that the rating can be chosen just like the force of the spell. This I think has had an impact on view of this method.



Thoughts to rectify this:
Make stealth an opposed test.
Change the roll to (attribute+skill+program rating).
Change back to (skill+program rating), but cap the dice gained from the program rating used by the logic.

FInal note: The Matrix is my albotross in this game. I can run it quickly-but not challengingly enough. Or I can make it very rough/imposible on the player, and slow the game down.

It is, actually. While the most common use of stealth is as a Threshold for the node you're hacking to notice, the SECOND most common is that its an Opposed stealth vs analyze test if someone wants to see you on the matrix. Like. It already is. Also in unwired, its referenced in other things like Spotting Traces.

I've slept a bit, and thought it over.

I think the matrix system doesn't really need more caps or complications to achieve realism; it's way too complex already. I especially think program options are awful because they tend to square the amount of programs you should choose from. I like the idea of abstracting all those options into the general rating of your program - if you want to use it in a weird way, take a DP penalty. A good program has more dice to "feed" to special "maneuvers".

Then, I think another point is this: are script kiddies bad?
I think it fits with SR's dehumanizing setting that S-Ks and Agents and AIs threaten to replace hackers; cast it a bit more film noir, where they're the last proud few. Only some hackers are good enough to still win from the machines.
IIRC, Neuromancer used themes like that a lot; the protagonist gets a spectacular icebreaker software to do his job, and it's all about freeing up the AI to enter a self-improvement loop.

OTOH, it might not be what you want to play. There are more plausible future possibilities that aren't necessarily fun. I don't like biometrics very much; it should be easy to compromise the anonymity that protects shadowrunners. Things that seemed survivable in the 80s aren't anymore; the idea of total surveillance/total market research is much closer to reality now.


Next up: "by your own brains" vs. program-dependent?
Modern computer technicians still use command-line interfaces, but the software behind that command line has grown a lot. The things you can do with a single short sentence are much more profound with modern high and very-high level languages. Coding in machine code or assembly is really rare nowadays; only virus programmers and driver programmers use it with any regularity - because it allows extreme optimization. And even they use software to aid their coding.
Modern compilers are pretty good at optimization, and with sixty more years are likely better at it than humans. For large, complex programs, doing it "by hand" is likely less effective; more mistakes will be made. There is a limit to how big a picture humans can work with, that computer processing power can bypass.

I think it's fairer to say that the quality of your programs and the skill you have at using them are decisive in the matrix. The problems would go too fast and be too complex to do "by hand" - the analytical powers of 2070s software are far greater.


In the end it comes down to what you like; but I think I made a case for the plausibility of RAW.

That's actually not true. Compilers are good at general optimizations, where a technique improves the efficiency o a whole class of usage, but its not good at unique, almost trivial, optimizations that are specific to how a programmer writes his code and what he wants it to do.

You are right of course that programs are "doing most of the work," but that isn't sufficient. Programs only do what they're programed to do, you need a brain behind it telling it what to do. My experiences from making maps for Left 4 Dead and Team Fortress 2 have made this point painfully clear:

While vvis.exe does calculate what volumes can see what other volumes (and vrad.exe calculates lighting) I the mapper have to know what to do in order to optimize those processes. Not optimizing means that a 5 minute compile time bloats into a 5 hour compile time. Time it takes to make those optimizations: About 30 minutes.

One of the best sayings: "Do not carve" and "Use lots of detail brushes, especially for complex brushes."

Creating new programs is still a Logic+Software test, so indirectly your mental stats still figure into the whole thing. And the sort of optimization you suggest is hard to do as fast as the Matrix would require it - in 60 years, computers would be able to do it as well or better than you.

Also, how to balance using mental attributes? They already do a lot;

Logic is used in improving your software and hardware. Not as brutal as the others:
Intuition determines to oh-so-important Matrix Initiative - does it need to to anything more to be good?
Willpower: biofeedback survival. It's already crucial
Charisma: hard to justify using it => Logic and Intuition will dominate almost all Matrix tests

I think you'd end up making mental attributes far too decisive for hacking, and way too strong. You'd end up in the situation that a hermetic mage is almost instantly a hacker too because he's likely to be smart. (A reverse script-kiddie; where someone with raw intellect forces his way into the net.)



I personally tend to think of Logic as a "slow" attribute; to exhaustively study and analyze a subject. It should mostly have uses that take a lot of time. Maybe instead of having a gazillion additional program options from Unwired we just need a "Take Aim" equivalent action where you spend actions to add Logic dice to any Matrix test.

Most importantly, the software has to be there in the first place. A command line is just another tool to run programs and specify input data for them, it does not magically make things possible which could not be done with a GUI.
If your computer does not have a packet analyzer software (GUI or not), you can't analyze network traffic no matter how much you hack at the keyboard. Sure, with enough time (like a decade or so) a good programmer with lots of knowledge about networks can write his own Wireshark from scratch, but who is going to do that? In theory (and SR), you could even go completely without program and just translate the electrical signals on the wire into this page but again, nobody will do this because it would take ages.


Depending on how you define "better" compilers already are far ahead. I'd say that if a compiler needs two seconds, while a human needs two hours and reduces the code by two or three instructions, the compiler is the clear winner (IMG:style_emoticons/default/wink.gif)



/signed

Actually based on the attributes not really. Look at astral space. Logic takes the place of agility (which is practically never used in astral), intuition takes the place of reaction (and is the sole stat used in initiative.. unlike cyber, and assensing/perception again unlike cyber). Then it goes awry because so wilpower is THE stat in astral. So you end up w/ Wil + Int being the two big stats (w/ cha taking the place of str) in astral.

We can agree to disagree, just I think we have a fundamental disconnect on what the attributes actually do.
Logic seems to be the stat for the initiatior... intuition the stat for the reaction.


Actually I was a little curious for your feedback on those ideas. I'd been bouncing them around for a while but haven't done enough w/ decker-riggers to be fully comfortable w/ suggesting things outside of ohh this sounds neat (I'm trying though... just seems the kiss of death whenever I roll one up).

SR4 is first and foremost a roleplaying game. If you want 'hacker' to be an actual archetype that people play, Script Kiddies are unquestionably bad because they invalidate actual players. There is absolutely no argument you can make which features playable Hacker Characters and also has Script Kiddies.

If one of the players in my game can be replaced by a box with 'Tracebuster' on the side, the game is terrible. Saying otherwise is saying you're happy to let one guy play Super Smash Bros while the agent does all the work, which is dumb because we're trying to play an RPG, not slamming Princess Peach with a Mallet (fun as that is)

Wireshark isn't actually a decade program =P but the points are good.

Speaking as a CoE (yeah, yeah, appealing to authority on the internet (IMG:style_emoticons/default/spin.gif) ) logic is less important than alot of people think when it comes time to actually make things happen. - it's *critical* in the making a tool part, but when it comes time to use that tool to make/break/analyze something else, the quality of the tool matters alot more than whether or not I know exactly how the tool works, and my general knowledge of how these *sorts* of tools (hacking skill / computer skill ) matters more than how much I know about how programs work in general (logic). So I defend the system as written.

I agree that it's weird to have logic 1 hackers being BP 'optimal' - but they're not gameplay optimal because of things like program degradation etc. so it's largely a theoretical problem for most table tops. (it is a real problem for one shot cons and some missions groups etc. but house rules won't help there... )

so uh, vote for the current system being fine, really? just don't approve characters with logic 1 who are primary hackers.

The other issue is that there are no rules for entering the game with a loadout of self-programmed programs. You either start with store bought ones (to get the job done) or none at all.

Been stewing over this for a few days, want to throw it out there.

Use Stat+Skill in Cybercombat only. Matrix attack uses the Attack program in both the attack roll and damage value. Instead use Logic+Cybercombat for the attack and the program rating for the damage value only.

And change defense in Cybercombat. Response + Firewall is always going to be the highest anyone can run in a node, makes Cybercombat loooong. Change it to Firewall only, add Intuition for Full Defense.

IC and Agents roll their ratings only and can't go Full Defense. This makes IC less potent, but you could always attack with multiple IC.

So Agents are basically bad because they replace PCs, and so would a measure of script-kiddyism be bad. Not because it's all that unrealistic; a guy with 80BP invested in Cracking and Electronics skill groups isn't really a script-kiddie in my view - that costs a lot more than Logic 5. But because it takes the stage from the player. Some possible solutions:

- Agents need a lot of supervision and skill to handle, like Spirits.
- Agents can be Tracked back to you, and can receive Spoofed orders from other deckers.
- Software degradation separates the elite hackers from the professional users from the script-kiddies. Even self-written software degrades, but anyting that's on the market degrades really fast because everyone can analyze it and find its flaws.
- Patching your software shouldn't take too long; serious high Logic+Software hackers should have little trouble keeping up.
- If a player tries to build a Logic 1 hacker, smack him with the Rod of Powergamer Smiting. You don't prevent powergaming with additional rules, but by dealing with the person.

What we need rules for is the following:
- The effects of online hacking communities sharing information
- An easy, low-bulk but sophisticated system for software degradation
- Limits on Agents, realistic but effective
- Rules for starting with self-written software/hardware
- Rules for decks bigger than and more powerful than a commlink to be bought in the store (big oversight in the books I think)

Did I miss anything?

that's the other point. you're already sinking significant BP into the actual *skills* - you're not a script kiddie anymore.

being bad at math doesn't stop one from being a genius in their field (ask any number of theoretical physicists who never quite got the hang of decimal addition for instance.)

Another argument against compilers v. metahuman/other hackers goes to the nature of the matrix itself. The matrix is a far more interactive, intuitive environment than simply running a computer or monitoring what comes in through an I/O port. Compilers, no matter how complex, can not deal adequately with a novel situation, of which almost any situation in a node with a conflicting reality will really be.

The basis for this is well supported in the fluff/history of SR. Causing agents problems whenever they are in a reality they are not specifically programmed for appeals to me. Just as running software on the wrong OS, its possible, but it takes additional programming, which the agent is incapable of (probably). Which the OS is not a very good example, but I can not think of a better one at the moment.

This would help make agents better for certain things, like passive defense in a node, and make them far less viable for actually hacking into another system, which is a compromise I like.

My experience with math software designed to be able to do calculus gave me that thought. While the software can eventually get the answer by using brute force methods, taking advantage of the increased processing power, humans are able to get a far more simple, elegant, and workable solution.

having gone through multi-variate differential equations (another appeal to authority! (IMG:style_emoticons/default/frown.gif) ) while the human may get a more 'elegant' solution after a lot of time consuming work, the computer will still get the right answer within expected tolerance, much much faster.

and ultimately, when it comes time to get inside that node, it's the right answer that's important, not how you got to it...

punishing agents for being in paradigms that they aren't designed for is good; but remember that the representation a decker/hacker 'sees' is *not* the underlying architecture of a system. agents would not interact with that representation except as incidental in the rendering of their interaction with the actual architecture.

just like a bot crawler from google doesn't actually bother to render images on the page when retrieving it for categorizing, an agent isn't going to bother to render the 3d display of the underlying data. - that's for human eye convenience and not for operational speed...

When you and I did ODE and PDE we used a pencil. Part of the backstory is that our minds are freed from slow clunky interfaces.

I don't know about anyone else but my brain can do any maths countless times faster than I can write out the steps.

BlueMax

What you see in the Matrix is a metaphor, changing the metaphor will not change the underlying system. Otherweise, every program would have to be reprogrammed for every host (IMG:style_emoticons/default/wink.gif)