Otaku required technological interfaces, could use cyberware, and eventually lost their powers. Technomancers are none of that.

Right, but SR ignores the rules of math by making decryption super fast.

How is that a fundamental shift?

"Abba-zaba... you're my only friend..."



Kind of like DES in the 1970s? You know you can brute force that faster than you can bake an apple pie, right?

-Wired_SR_AEGIS

Technomancers are irrelevant. If they are communicate with a purely electronic technical device, which the Matrix is described as such, then they must be doing so in a manner which those devices understand. It doesn't matter if they're depositing electrons into their received or simulating the movement of electrons along the circuits. If those electrons create a signal the device does not understand they're not communicating and not hacking. As far as I'm concerned Technomancers just have a biological network interface. They may not be a node themselves but there must be nodes present for them to interact with. A technomancer cannot hack a remote node that isn't connected to the Matrix.

Let's look at it another way. You could destroy every link for the Matrix that crosses the oceans. Undersea cables, satellite uplinks, whatever. The matrix will still exist and still be the same even though the Americas have been cut off from the rest of the world.

--



Decryption is always super fast if you have the key. That's why P=NP being proven true is a problem for cryptography. Every bit of encryption is quickly broken if you have the key and thus the solution is quickly verified. However solving the problem of what the encryption key is currently takes a very long time. If cracking encryption is fast, which Shadowrun has dictated, then in the world of Shadowrun P=NP has been proven true. There's no defiance of mathematics since P=NP is itself an unsolve mathematical theory. No one knows the math to quickly break encryption, if it exists. That's the key though. We don't know if it exists or not. The other thing that exists in Shadowrun that shows that P=NP is the prevalence of highly advanced artificial intelligence.

Well it's like saying street samurai no longer need guns because they have mind bullets.

Duh, that's why everyone plays Mages!

StealthSigma is correct: Decryption is always super fast if you have the key. The parts that take a long time are various cryptographic attacks (The longest of which is a brute force attack) and/or asymmetric encryption.

I think in Shadowrun Encryption you'd need some kind of one-time pad for secure communication, which is mathematically unbreakable, and the attacks would have to be some sort of side channel effort. Or maybe associated with attacking the key distribution. Except... Key distribution itself becomes a fairly fundamental problem.

I'm not sure how hard it would be to attack. Do you, like, just walk up to someone and hand them a key? Because you definitely aren't sending it over to them if you intention was communication security.



Ain't that the truth!

-Wired_SR_AEGIS

Personal Opinion... Not fact.

Decryption without a key in SR is extremely fast. So it proves P=NP in the SR universe. Key distribution doesn't matter because the encryption can be broken quickly. Since encryption can be broken quickly I can think of not reason to use encryption on data that is at rest. There might still be some value to using encryption on data in motion but those are mostly hinge cases where the amount of data being transfered is small and the window of time it needs to be protected is much faster than how long it takes to break it.

"But," you might say, "the processing power on devices has really gone up since modern day so they can break encryption faster." This is partially true. Modern day encryption will be broken faster by increases in processing power. However, those same increases in processing power combined with the increase of data transmission rates permited even more complex encryption keys to be used without causing a performance decrease. With P=NP you could do a 2,097,152 bit key and it would still be solved as fast as a 1024 bit key.

Indeed. Which is why I specifically mentioned a one time pad in the context of your discussion on a world where P=NP.

For every message encrypted by a one time pad, there exists any number of messages which decode to the same length. So P=NP isn't something we're concerned about because attacking a perfect one time pad isn't a product of computational power. What we're concerned about in that paradigm is secure key distribution for our one time pads. Or protection of side channel information related to the ultimate one time pad implementation.

Note: All that said, I'm not comfortable with saying that in Shadowrun P=NP. More likely I suspect that the intent is that the implementation, not the underlying algorithms, are susecptible to attacks. Which is why hackers experience variable success against varying degrees of hardware implementations in-game. I was just following your logic along for a bit.

-Wired_SR_AEGIS

Ah, I think I understand where the disconnect is now. Please correct me if I'm wrong here. You seem to be of the impression that technology in the Shadowrun Universe, has had (effectively) a linear technological progression from SR1 to SR2 and so on through SR5?

From what I've been reading from the dev posts and seeing in the previews, SR5 is a complete reinterpretation of the underlying mechanics of the Shadowrun Universe. So smartlinks didn't grant a +2 dice bonus (SR4) then become stupid and start granting Limit increases (SR5). (With a +2 dice bonus if you're wirelessly connected.) In SR5, they've always granted Limit increases from the day they were invented. Once they went wireless in 2065, they started gaining the bonuses. From 2065 - 2075, you couldn't turn the wireless off, really. Route it through your commlink for security? (Skinlink or no.) Sure, but it's still connected even if you're running in hidden mode. In 2075, now you can turn it off. Going back to smartlinks from 2050 isn't going to help you, in universe, because there never was a mechanically different version.

Call it a retcon. Or a reimagination. Or yes, even an ass pull. For a parallel: In D&D2e, you had THAC0 and quite limited AC to hit ratios. In 3.x you had run away attack modifiers with equally impossible AC. Same worlds, different mechanical interpretations.

For the record, I do see some of the bonuses as ridiculous. At the same time, baby with bathwater here. Mechanically speaking, this rules change doesn't worry me nearly as much as some of the others do.

-Temperance

Be Nice.

People are not going to agree with each other here. They won't be convinced by your arguments.

Say your piece, accept differences, move on.

Spamming behavior is also against the ToS.

Think about how one line(or wall of text) non contributory posts in every third or fourth response in a thread qualifies as spamming.

Stop yourself.

Except the Matrix wasn't wireless until 4e circa 2070ish. Prior to that Smartlinks couldn't be wireless because Cyberdecks weren't

That doesn't change Temperance underlying point, which is belief that that the Technology retcon indicates that Smartlinks in 2055 gave bonuses to limits. Not bonuses to dice.

-Wired_SR_AEGIS

Is it a retcon? Maybe. But if that's so, a lot of stuff was retconned from SR3 to SR4. Either way, as far as SR5 is concerned a limit bonus is all a smartlink ever provided until the new new Matrix came along and better functioning became possible.

That's as maybe, but it doesn't get around the fact that +Limit as a mechanic fails to make a Smartlink do what a Smartlink is supposed to do, which is make hitting a target easier.

Yeah, but if they hadn't done the wireless bonus thing +Limit would have been the ONLY effect of the smartlink in SR5 - so that's not something to blame on wireless bonus. We know from the design blogs that they first decided to take out a lot of the +dice stuff, and later came up with the notion of wireless bonuses.

Totally RHat. But as was mentioned earlier or elsewhere the more logical Smartlink layout is reversing the benefits, making it do what it's meant to do with +dice as standard and then +limit for online (then it works just like Section 9's Smartlinks from GitS). It makes an average marksman good and a good marksman great if using the matrix targetting info.

That might be one way to go for houserules, certainly. Helps everybody, and the wireless option provides something that the skilled shooter can make use of. Just contrasts the design intent (reducing the +dice bonuses). I'd say the Matrix bonus would come from being able to use more sophisticated processing to account for more factors, both externally (like wind over range) and internally (correcting for imperfections in the weapon itself). Connections to additional sensors would provide some assistance as well, such as identifying relative position to friendlies or bystanders to avoid crossfire or hitting someone with a round that passes through the target, and to predict where they're going to be when the bullet arrives (accounting, specifically, for the time it takes to pull the trigger and for the bullet to travel), thus showing you precisely where to lead the reticle.

Now, the bonuses of the extra processing power sound to me more like they would add dice, but there's also a further method you could go for in a house rule if you don't mind making smartlinks more powerful: Standard is +2 Limit AND dice, connected is +3 or +4.

I think something that was forgotten in the discussion about encryption, is that a DNI or direct link does not need encryption. Rather than taking (N + W) ms to travel across the wireless and back, it takes N ms to travel across a direct connection. Unless N is a non-euclidean number, it will always be less than (N + W). If it IS a non-euclidean number... well, just have the mage counterspell it, or you can kill something to create a background count and render the technophiles useless. If the technology is working off of magic, then mages are even MORE powerful. If it's based on science, it conforms to a relatively strict set of rules, and if the data is still traveling in non-euclidean pathways (not including Resonance-active entities), then there's much more profitable things to be using it for than matrix security (lottery, extra-dimensional communication, etc.).

The science behind matrix topology HAS been provided and explained, the way it works on a fundamental level has been explained. The reason for the quick adoption of Matrix 2.0 was because of the giant crash of Matrix 1.0 (Revolutions), and the expanding use of wireless in our (present day) lives. Trying to explain to people with the internet at their beck and call and technological insight enough to pull up such resources, not to mention the simple principle that running down the street to get a key and running back will never be as fast as opening an open door right beside you; that somehow running to the end of the street and back for security protocols is faster than a secure, direct access to the device without the need for security protocols is misinformed at best and disingenuous at worst. Handwavium can't change that simple principle.

We're not even arguing about ALL wireless bonuses, here. We're just puzzled that a simple, fundamental principle that governs all knowledge of science and reality, outside of theoretical time-travel (magically and technologically IMPOSSIBLE in Shadowrun) and warping/dimensional folds (also impossible technologically and magically outside of the 4thEd Levitate spell in Shadowrun), EVEN governing magic, is thrown out the window entirely for some simple, non-game-breaking facets of technology, and, granted, some powerful bits that logically should also have the option. Considering that the conductivity/transference of air is nowhere near as fast and efficient as metal wires (copper in the modern) or fiber-optic cords (substantially less impurities, direct line-of-communication provided) or nervous tissue (comparable in basic function to metal wires), there's outright no conceivable way, even in the rules as presented (no teleportation or time travel) that matrix topography would be able to outpace a quick DNI/Direct Linked command.

Encryption is a burden for wireless and not direct connections.
Time travel is a non-issue
Teleportation is a non-issue
Computational power can be substituted on-site (on-device).

And those arguing against Wireless bonuses are still willing to accept a fair few as reasonable and easily explained. It's not "all or nothing." Compromise can be reached (and probably will be via house-ruling). Like the XBone's initial presentation, it's a neat idea, but poorly executed. There's no way a corp wouldn't be ready to reap the benefits of AAA's like Horizon and Ares scewing the pooch, boosting themselves up to world-power status with All-secure Non-Wireless SecurTech Augmentations and Enhancements.

Even irl I favor wired connections (stronger, faster, better) over wireless (limp).

I think too many bonuses was what they wanted to avoid.

I agree with the idea of limiting dice in general.
Limits are a cool way to reduce the net success over kill so common is SR4 and provide bonuses to gear and cyberware that doesn't add extra dice.

But I think they took it too far with Smartguns since a SR5 smart gun is almost worthless to someone with limited skills.
I mean you have to be somewhat skilled in order to need the limit increase at all.
Some cyberware should still add dice.
And if you are fairly physical to begin with then you need to be really skilled before a non connected smartlink is any use at all.

It makes much more sense for the smartgun link to add dice so it helps everyone out no matter what their skill level.
Especially if all you want to know is exactly where the gun is pointing at right at the second you mentally command it to fire.

I see standard smart gun features helping to put the bullet on the target and hitting it (put the dot on the target).
Anyone can hit a man sized target in the chest if there is a dot showing where the gun is pointing.


Then if connected to the Matrix, the smartlink should adds to the the limit and aid the experts to be even better.
The extra processing power and data being connected would let you put the bullet where it will do the most good on the target.
It takes an expert to put the bullet through the gap between his shoulder guard and his chest plate.
And that is where your extra processing power, range finders, and data streams would make the difference.

One wireless bonus I could see is making smartgun effects reduced at long range (say double the ranger where a laser sight starts to lose effect) unless you have ranger finders or a wireless connection.
The further away you are the more bullet drop, the more time the target has to move, the less accurate the weapon,...
The more variables that a wireless connection would allow the smartlink to compensate for.

I especially want the added dice as a GM, I get a lot more use out of a smartgun link than the players (adding two dice to a ganger with a skill of 2 is a much bigger boost than adding two dice to a runner with a skill of 5 and a specialization.

The SR5 Development Leadership is making life so difficult for the SR5 GM!!!! (in a few things!)

Good points all.

BUT I LOVE EVERYTHING YOU JUST SAID!!
Smartlink offering dice offline, Limit boost on Matrix, cool! That even helps give professionals something more to consider. My decker/biotech is STILL going to be researching/producing modules to eliminate that need, but it would place such a thing a bit lower on the burner.

That's true in a number of cases. A great many number of cases, actually. Frequently with discrete systems that are subcomponents of a larger system, there are very few reasons to employ some cryptographic component. At least today. Such a component isn't related solely to the privacy of data, either, as (I mentioned earlier) cryptographic knowledge plays an integral role in authentication.

However.

Whether or not there is a 'need for encryption' is really a question of implementation, not a blanket quality of directly connected devices.

Granted in the case of DNI, the path should go Brain->Wire->Device, you may be able to build a strong case for not needing an authentication method because you are operating under implicit trust.

However, first you must assume that despite that direct path there is no need for authentication. In 2013 that's frequently true. Is it still true in 2075? It's a world where people manipulate the matrix with their mind, afterall.

Personally, I don't think it's completely unreasonable to believe that there's overhead necessary even for direct connections to establish trust. Especially since we've already handwaved so many elements of "science" to get ourselves here to begin with.

It's like, we've hiked the grand canyon. Why not go another yard?


So, to be clear, the question is not a matter solely of time. It's a matter of time as a function of processing power. The case being built is that the cost of W is dwarfed by the cost of that processing power.

So the place that your argument will be most effective is not network latency, but whether the fundamental need for distributed computing itself is valid.



Again, if that's an argument that someone is making they can probably put it to rest as clearly it isn't going to make any traction. It certainly isn't an argument I'm making.

-Wired_SR_AEGIS

The predicament, though, is that I MUST NOT assume that. In fact, the direct connection and DNI implies the exact opposite: that authentication is NOT needed. People manipulate the matrix with their mind? That's an even better reason not to put your vital systems ON The matrix.


This is a valid point for SOME ware/gear and its functionality. I just stated that this isn't a blanket voiding of the need for Matrix connectivity, per se. The point I'm making is that for a command as simple as "close" or "open" through a secure, direct, unencrypted, unobstructed channel is, by all rights and by all reckoning presented in shadowrun, barring outright lying about the rules on the part of the people PROVIDING the rules, always faster, and based on fundamental principles of the world and how it operates. While I might consider a rule or two poorly worded or maybe not entirely thought through properly (ala XBone's marketing and previously intended DRM practices), I'm not quite ready to consider them to be lying to us about the rules of the game they have themselves brought to us.

In the case that wireless signals DO defy the laws of time and space to make simple commands (such as Open or Close) travel faster, despite not needing that much processing power, and needing authentication and encryption to properly function over the wireless mesh network,...
1) this means the Devs have lied about, or at least exceedingly obfuscated, the way that time and space functions in Shadowrun, and
2) my runner is going to say "shove it, I'm goin' gambling" and making a tidy sum off of sending himself advice from the future.
I don't see either of those being rational, logical, or economically viable probabilities.

However, it IS what you're discounting in the arguments that you make. Conductivity/energy transference is the core of the Online/Offline bonus discussion. Saying "Magic" means a mage will just dispell your technology, thus rendering all technology impotent in the face of magic. Saying "handwavium" means that you have no clue how to support your argument, argumentum ad magice or argumentum ad nihilum cover it (arguing on the basis of magic, or arguing on the basis of absolutely nothing to support your argument). In the face of arguments based in science fact, observable, rational, reproducible instances of physical law, topical expertise, etc., an argument based on "it's magic" or "I have no support for my claims" fall flat.

For another example/analogy:
Say we both had the goal of moving a vase in a hotel room from the wall nearest the window to the wall nearest the door. We're both given that task at the same time. I'm standing in the room right next door, with a direct, unlocked door between the two rooms. You, meanwhile, are standing outside the door, and not only do you need to get the key, you also need to solve a complex mathematical problem while you do. You'd have to take the elevator down to the front/concierge desk, asking people along the way to help you out with that math problem, take the elevator back up, input the answer to the math problem, and granting that your answer is right you can THEN use your key before you could handle that simple task. Meanwhile, I just have to open a door, walk over to the vase, move it to the other wall, and I'm done. Who will be faster?

The other hand, the cases I'm NOT arguing against Online bonuses for, is as follows:
We're in a hotel, same as before. I'm right next door to the room in question, same direct access. You're outside that door, same need to answer a complex math problem and get a key before you can enter the room. This time, however, is that there's a simple, unhackable interface in that destination hotel room, and the goal is to solve a SECOND complex math problem as fast as possible. I'm not allowed to leave those two rooms, I have no internet or special tools (I get pencil and paper). You don't get any special tools, either, but you can talk to people who DO have special tools or internet access.

So, for a simple command (Open, Close) or something that doesn't NEED to compute, I don't see this being viable, rational, or acceptable.
For, say, smartlink (complex algorithm to compute trajectory based on airspeed, wind, thermal, etc.), I can dig it.

to everyone: sorry for being so verbose... or would it be overscriptive?

The biggest issue I have with the wireless bonuses and wirelessly Meshing with the devices around you to connect to the matrix concept is very simple.
Every time a runner connects wirelessly to the Matrix he is leaving a "Fingerprint" on every device his gear meshes with.
Likely every device in a kilometer or two radius.
And by looking at what and how his devices are meshing to connect to the matrix, you can tell exactly what the runner was doing.
And with that many devices, there is no way to clean it all up.

It is not like you are a decker using fiberoptic to and spoofing your location (so there is no meshing involved)

I mean you go from 4th where there are huge sections of fluff on how to kill all the RFID that are in your gear, to SR5 where runners get wirelessly connected to the matrix all the time and it is no big thing.

Remember of course that

A) SR4 was awful, and much more importantly

B) Shadowrun, to work as a game system, has to be systemically stagnant.

To clarify on B, you can't progress the game world 30 years and follow any sort of realistic path of technological growth. Otherwise strength enhancements would let you jump to the moon and smartlinks would be able to predict up and coming Lone Star agent superstars as they're being born and shoot them while they're in the little plastic boxes at the hospital nursery.

That is to say that every single time we get a new edition of Shadowrun, the only consistent rule is that last edition's bleeding edge is probably just too extreme to base the new core technology on, and so we get the original stable of tech with a shiny new veneer.

I'm pretty sure that laser weapons were a thing in 2060, yet no one seems to have them 15 years later, which is about the time it took in real life for laserdisc to become an incredibly awesome new thing and then quietly go out of style.

Why? Because if Shadowrun actually built off itself from edition to edition, the technology would go right off the scale and then nothing would be any fun anymore, because you'd have nothing but genengineered biorobot monkey laser drones with magical armor blasting through Aztechnology facilities to acquire the latest in soychip formulae.

And nobody wants that.

Opinion, one not shared by many, and also irrelevant to the topic.



That last part is sarcasm, right? Cause personally, that sounds awesome.

If Shadowrun built off itself in each edition, we could expect to see all three areas advance - 'Ware, Magic, and Matrix all moving on forward. So long as they kept pace with each other, due to the specific way in which Shadowrun is balanced, all would be well on the game end of things. The setting, however, would be RADICALLY altered.

I’ve done my best to read through all of this thread, it’s not an easy read. Let me just say I’m not a huge fan of all of the “Matrix Bonuses” as they are presented, I do think there is room for improvement.

Now I’m not saying this is right, but what if the switch to everything being matrix connected has less to do with a change in the fundamental hardware and more to do with a change in protocols. Here’s my though-process on this:

In SR1-3 (more or less) you had basically 2 protocols in use; there was the matrix protocol (MPv1) and the direct neural interface protocol (DNIPv1). During the switch from SR3 to SR4 a major world-altering event happened (Crash 2.0) and caused a major shift in the way the Matrix was accessed. This change caused a switch to the now existing wireless Matrix (MPv2), DNIPv1 is still in use at this point as well. So now years later the next big event happens (I think I read somewhere that something bad happened in Stormfront). This major change is forcing everyone to switch to MPv3 due to its security, perhaps during this same time there was a flaw found in DNIPv1 as well, and rather than creating a new DNI-protocol the corporations in their infinite wisdom simply said “Well drek, we just made this MPv3, why not use it rather than investing in a new DNI protocol?” and thus the universally hackable on-line everything was born.

This is an easy way to explain things should only be connected to your PAN being effectively connected to the Matrix at large and why things that used to be strictly internal (using DNIPv1) now being accessible by the Matrix.

I know this may sound dumb, but at least it’s an attempt to explain some of the silliness. It may not explain everything, but it might help on some of, if not most of the bonuses.

It does not imply such a thing. It may be implemented in such a manner, and it may not be implemented in such a manner.

You may certainly infer that a DNI has less of a need to operate outside of an implicit trust environment than a mesh network, and I may even agree with such a conclusion. You may not, however, make a blanket statement that such a mechanism is not needed because even some systems today with direct connections require authentication.

In fact, if the trajectory of today's technology is any indication authentication is becoming more and more and more and more important throughout the numerous layers of systems architecture.



Right, so again, no one is disputing that.



Right, so again, no I'm not.

Conductivity/energy is certainly part of the discussion. But it is not the definitive end-all-be-all. For a number of reasons. And even so, in the real world today some directly wired protocols offer more overhead than others when implemented. So even inside the world of direct connectivity it's not a question solely answered by physics.

Once again: Implementation



Sure. Here's another scenario. Whoever can have the full contents of the library of congress read aloud to an audio sensor on the door can pass through.

You can try it by yourself. I'm going to get my friends to help me.

Let's see who passes through first?



You're making assumptions on what is 'simple', and what is 'needed' which are important to recognize as assumptions. There is no fundamental requirement that your assertion is correct. It may be correct. And it may not be correct. But based on the Matrix in 5th edition, it appears that they are not correct.



It doesn't sound dumb, DMiller. It sounds highly plausable. Standardizing, and unifying communication protocols is a pretty big deal as it turns out. Less interfaces (Logical or Physical) to implement and maintain provides more manageable attack surfaces for a system, and adhering to an industry standard typically provides a more robust, and secure final product.

-Wired_SR_AEGIS

The problem in that analogy is that, with the advent of a "Standardized and unified" communications protocol, rather than creating more manageable attack surfaces, the number of attack vectors just exploded exponentially. That strict adherence to the new standard did not create a more robust, secure system. It created so many more holes that it is laughable.

Are you talking about in real life? False.

Unifying standards leads to more rigorous investigation of the underlying limitations and assumptions of that standard. Such scrutiny is essential in the world of security. And it is the lack thereof, best demonstrated by proprietary standards, that frequently leads to the exploitation of a system.

Are you talking about the world of Shadowrun?

Game mechanics drove the introduction of attack surfaces. They arrived there because they decided it was beneficial to a character archetype. Then supported it with technology. Not the other way around.

-Wired_SR_AEGIS

Rules handwave around 80%. You have to accept it. If you go full realism a lot of things would be different. Every part of cyberware might have its special "loophole" even depending on who build it. May it be a factory build in Shutdown command and emergency trigger etc.pp The point is, thats not a lot of fun to read and to remember. So you need in some degree to simplify it.
The 4E tried to say: Alright you can turn cyberware offline but you do not get the benefit from beeing online, but you can't be hacked...
Now what happened: People started to claim their cyberware was linked inside their body, never having paied for it or having any drawback.
As long as you had a datajack I could swallow it at least with headweare. But with I have a trodenet and it is going over the trodenet etc.pp it just started to feel silly.

Now they went the way of simplifing it and making a real zero and one case out of it. As I pointed out I would have liked going into the other direction, but I can see why it wasn't done. (I am also one of those guys who loved what project nevada did with the rules for Fallout New Vegas implants to a point)

This would have also been a great idea to make cyberlimbs a bit more appealing.
But none the less, this roade was not taken.

The problematic part being that the presented technology quite obviously breaks "suspension of disbelief" for many people. And while I certainly appreciate the efforts of forum users trying to present "logical" explainations, I still can't see how or why a point-to-point connection with a physical layer called "wire" (in a world that knows superconductive materials at room temperature and above) could be outperformed by one that uses a physical layer called "electromagnetic wave" and I certainly cannot see how or why any protocol implementation (which resides above the physical layer) that used either form in a direct point-to-point connection would be slower than a connection that has to be routed over at least one other end point ... unless of course you're now aiming for an explaination where the protocol itself deliberately slows down data transfer for a non-routed, direct point-to-point connection. And once you go there, we're right in a world where deckers can succesfully hack the GOD-controlled Matrix protocols for any connection and even on remote enemy gear but fail to do so - while using the the very same protocols - whenever "friendly" personal gear is involved.

The catch there being: The makers quite obviously trying to shove their idea of combat hacking down the throat of a player base despite the shown dislike for the previous inception of the same idea. And to make things worse, they can't even open up that part in advanced rulesets without making themselves "fools", because once such a means officially exists, their shiny "new" toy of "combat hacking" will be rendered useless once again.

@Chochise

Right, so there are a number of things that come into play above and beyond simply data transmission rates that may introduce overhead. So getting hung up on how fast electrons shoot down the wire is missing out on the whole story.

I mentioned authentication above:

Device1: "Hey, Device2. I need you to perform command X."
Device2: "Hey, Device1. I acknowledge that you would like me to perform command X (Note: This may not be the best way to structure authentication, but is easy to follow as an example.), would you please prove to me that you are authorized to issue that command."
Device1: "Yes, Device1. Here is my authorization."
Device2: "Thank you, Device2. I accept your authorization. I will now perform that action."

That right there introduces overhead. But a number of other things may as well. Understanding that I'm speaking in general terms, error checking can introduce overhead. Or at a basic, jr. Network Engineer level, Protocols that are connection oriented vs. connectionless protocols carry varying degrees of overhead. Commonly implemented in distributed systems, that is by no means the only application.

Device1: "Device2, I need you to perform a command, are you available?"
Device2: "..."
Device1: "Device2, I need you to perform a command, are you available?"
Device2: "..."
Device1: "Device2, I need you to perform a command, are you available?"
Device2: "Device1, I acknowledge that you need me to perform a command, I am available."
Device1: "Device2, please perform: X."

So yeah, once again, implementation complexity plays a significant role in speed. Things as simple as checking to see if you're going to overrun your buffer (and thus prevent numerous nasty, real world attacks) vs. not checking, and assuming that you will never overflow your buffer have huge implications in the ultimate performance of your system.

Without understanding that system implementation, it's really difficult to make a sweeping gesture about how things "should" be working.

I'm not going to argue that, today, directly connected systems will largely operate with implicit trust and will dramatically dwarf the performance hits of some, silly, insecure wireless implementation of a similar function. But once again, that's based on implementations today.

...There is no absolute law of implementation that says it will always be that way, particularly if the cost associated with protocols begins to emphasize the latency of the transmission medium less, and emphasize the complexity of the underlying computing more.

And once again, because it's worth repeating, I don't buy into the full vision of the Matrix that the Shadowrun Developers have come up with. I just happen to think that there are much larger fish to be concerned about, if you're going for "realism". This is really just small stuff. Particularly in the context of a retcon.

-Wired_SR_AEGIS

Talking about the game.
And yet, your premise is that such a design implementation is good, becasue it is more secure. (Isn't that what you said?)
Mechanically, that is not what happened.

Not really. The only primary threats are illegal deckers with illegal programs. Joe with a commlink is unable to pull off any attack protocols anymore since the Matrix crackdown.

Ahhh, okay.

Well, here's another thing. All things are relative. So phrases like "more secure" and "less secure", need to be followed by the question: "...relative to what?"

DES was more secure, relative to 1970, than it is relative to 2013.

So, while I understand this may not be the most satisfying answer, we could say that while security may have improved on an absolute scale, on the relative scale of threat-->vulnerability, the relationship has lost ground to kinder, gentler, simpler times.

And... yes, Epicedion hits a good point. By volume, the threat landscape appears to be much, much, much more secure. These new protocols have all but terminated script kiddies, right?

No more downloading metasploit 2075 edition, and pointing it at your buddy's XP image,lololhackedyouwutwut!.

Whoops! Double post!

-Wired_SR_AEGIS

Except that the world at large is not really relevant to Shadowrunners. What was once secure for Shadowrunners is now wide open. Same goes for the Corps and their people. And THAT is where the mechanics break down. Corporations are wide open to Hacker/Deckers now, where as even in 4th Edition, there were things they could do to secure their stuff. Now, not so much. Corporations have made themselves MUCH more vulnerable to those that would be attacking their systems/personnel, and the mechanics make it so that they can be compromised with simple rolls (vs. the extended rolls of previous edition). Makes absolutely no sense.

And if the solution is to simply shut off Wireless, then the whole development path for 5th Edition is irrelevant, since the design goal was to have everyone online (so the poor bored hackers had something to do).

You can't compare those mechanics -- characters in SR4 didn't know what kind of rolls they were making relative to characters in SR5.

Two things happened:

1) The technology changed to cut out the enormous swath of readily available hacking, and

2) Hacking got faster.

What you're implying is that #2 happened because of #1, which is why you're annoyed. If you imagine instead that #1 happened because of #2, then things look a lot rosier -- every script kiddie could potentially wreck every device they came across in record time, so there was a general lockdown. Runners then found the cracks in the system and wedged cyberdecks in to keep hacking avenues open for themselves.

Try getting the name correctly next time, please



Overhead is a sympthom of protocols not physical layers. And my comments directly refered to the usage of the same protocols in either form of transportation.



Not actually understanding what I was talking about and then implying that I'm missing out on the wholes story is just two things: annoying and the opener for ad hominem. I sincerely request that to refrain from continuing with either.



Your example is a "protocol". No more no less.



And alll forms of overhead you mentioned so far would have identical overhead for Device 2, regardless whether the signal came over a different end point (the Matrix mesh) or directly from Device 1 on a wired connection ... provided that both Devices are capable of implementing the needed protocol in the first place.



On a serious, not so basic, senior Network Engineer level: The varying degrees of overhead between different protocols is a non-issue when explicitly comparing situations where the same protocol is used just with a different physical layer between two distinct end-points in a point-to-point connection.



Now thes example get's even funnier, because either you just did what I said you'd do, i.e. you introduced a deliberate delay mechanism within the protocol that causes Device 2 to slow things down, or you just made Device 2 having general signal reception problems. The former would again lead to the question why deckers/hackers would be incapable of hacking (thus altering) the protocol and it response time, while the latter would make Device just as unresponsive to attemps of communictaion coming from a different end point (a.k.a. "Matrix") thus not getting faster when compared to a direct input from Device 1.

Again you're just describing different aspects of protocol implementation ... which where and still are irrelevant to my general objection, since my "assumption" uses the same protocol regardless of used physical layer.



I have the sincere feeling that I'm not the person that lacks understanding here.



I can and will argue that even an implementation of future technology that removes things like implicit trust and other protocol differences, will neither change the fact that wired connection (particularly in a technical surrounding that knows super conduction at room temperature and above) will be faster than wireless connections that don't make use of quantum mechanisms (something not introduced in SR technology so far). At best the wireless transfer speed will be identical to the wired one (in a totally idealized situation).
I will further argue that as long all envolved devices are capable of implementing the required protocol, the direct connection will also be faster than a connection that hs to use more "hops" (at worst, again under implausible idealization they will have identical speeds).



Let's say that the known laws of physics of the real world as well as the laws of physics currently described for the Shadowrun universe are quite "absolute" when dealing with the supposedly involved physical phenomena.



So now you're (again) trying to tell me that the involved end points need more computational power / time when handling the identical protocol depending on which physical layer is used and how many hops the data had to travel (1 vs. 2+)?



I used the words "suspension of disbelief" with good reason. And just because you might see "bigger fish" out there to be concerned about doesn't change the fact that I'm worried about this particular fish ... just as others quite obviously are.



And my beef is not about its retcon nature, but about the missing constistancy within the game world ... where hackers are on the one hand perfectly capable of manipulating protocols on various levels but must fail to do so with end-to-end connections "just because" in order to maintain the concept of "combat hacking" viable.

To be explicitly clear, I agree with you that the latency induced by communicating through our understanding of a wireless transmission medium will exceed the latency induced by communicating through a wired transmission medium. So, it sounds as if we are in violent agreement in our treatment of physics.

Where we appear to be disagreeing is the necessity of also considering the mechanisms of the underlying implementation when calculating how we describe things like the responsiveness of a system.

I assert that a discussion of intelligent communication between devices without also making considerations to the implementation in addition to the transmission medium is meaningless. It is meaningless for the following reasons:

1) The transmission of "data" may have a means of distinguishing itself from noise.
2) The transmission of "data" may have a means of identifying errors.
3) The transmission of "data" may have a means of authorizing commands.
...
...
...
X) The transmission of "data" may have a means of obfuscating itself.
X+1) The transmission of "data" may have a means of reporting task state.

etc, etc.

Therefore, in your treatment of physics and physic alone in making your assertion, you ignore the the fundamental problem: We're not discussing shooting electrons around the world. We're discussing shooting electrons that mean something. This requires some element of computational power. Implementation, therefore, is an essential consideration. By definition.

-Wired_SR_AEGIS

how much computational power does it take to compute that it's time to close a vent?

and no, i don't buy that somehow you need to have secure authentication when going from your brain to a device located inside your body connected by a wire internal to your body.

if that isn't secure, then your damned computer needs to have a computer between every component to authenticate, and those computers inside your computer need computers inside of them to authenticate, and so on and so forth. eventually there either has to be some ability to make a simple device which is secure, or else nothing can ever be secure no matter what you do.

we know that taking your stuff offline makes it secure. therefore, there absolutely must be some capability to build a device that cannot be accessed via the matrix.

therefore, in a completely closed system, such as the one running from your brain through a wired connection to your cybernetic lungs, no authentication should be needed, because there is no way for someone to insert a command external to that system. it *must* be safe, because if that cannot be made safe, then nothing with a computer can ever be made safe no matter what; if i can somehow force a matrix command in between your brain and your cybernetic lungs which are only connected by a wire, then i can also force a matrix command in between two components of your computer which are only connected by a wire.

and even if it runs on the same protocol as the matrix just because nobody wanted to invent a new protocol, all it takes is a hacker to go back, grab the old protocols, and make them run on the new hardware. unless SR 5 actually retconned the existence of wired connections into something that never actually happened, and every single device is a bunch of free-floating wirelessly connected devices...

in which case we get right back to the point where nothing can ever be secure, whether online or offline, and we know that is not the case.

Actually not, since I'm even willing to accept identical transmission speed with the different media (a.k.a. "physical layer" or "layer 1").



"Underlying"? You can't go lower than physical layer. And to repeat myself: I was and still am talking about a situation where the involved transmission protocols themselves are identical.



And I will continue to assert that I don't have to consider the particular protocol further once the premise is that the protocol used is identical regardless of transmission medium and the transportation mediums themselves are at best (or worst, depending on perspective) identical in terms of signal quality to noise ratio ...
So now we're at a point where you're trying to explain some form of "meaninglessness" to me that was and still is "meaningless" for what I said/wrote.



Number 1) is a matter of signal quality within the medium. So unless the Shadowrun physics now suddenly - in defiance of what of "violently" tried to agree upon - make wireless transmission less subjected to noise than wired links (still under superconductive conditions!), that point really is "meaningless" as reason.
Number 2 and 3 are part of protocol implementation, which I have said to be identical for how many times now? So both of them are also meaningless as a reason for your attempt of explaining "meaninglessness" to me ... and that under the false pretense that I needed such an explaination in the first place.



Again just protocol elements where the involved devices need to be capable of handling the protocol in the first place ... otherwise they could not communicate with each other. And even if there was a limitation of local processing in a way that the en- and decapsulation of data within the protocol by miracle makes mesh transportation with additional computation power seemingly "more effective", you'd still face the situation that good hackers/deckers/engineers would or rather should be capable of bypassing the protocol requirements in order to transmit the same "message" when compared to the "full run" ... simply by using a less complicated protocol. If you really insist (as the current rules do from a meta-gaming level) that this goal cannot be achieved the very existance of hackers/deckers/engineers that are capable of hacking the SR Matrix in any shape or form is rendered into an absurdum.



This gets annoying as hell again, since I nowhere restricted my "assertion" to just shooting electrons around the world. So I simply cannot have ignored anything of that kind.



And the computational power is needed at both ends in order to even be able to communicate via the protocol in question. And that computational power does neither care for the transport medium nor the number of hops between sending and receiving end point. The computational power for decapsulating the pure data required to execute a certain command from within a transmission with a definded protocol by Device 2 in your example simply will not change just because Device 1 uses a one hop direct cable transmission or a several hop wireless transmission with the exact same protocol. And assuming that the local computational power is simply not high enough to compete with a distributed handling of said protocol still begs the question as to why noone would be able to just implement easier, faster protocols by themselves when the basic premise for hackers/deckers actually is that they can do so - even against strongest encryption - for every other part of the Matrix environment.



Not in this particular case. So please refrain from further attemps of providing "plausible" explainations to me for the makers decision to turn device communiction in some form of "magic" just to maintain the "combat hacking" concept without providing their own explanations, because frankly, that's their job not your's ... and within my perception you're not doing better than they are. So at the end of the day you won't convince me more to "buy" your explaination than "they" convince me to "buy" their product ... and unlike SR4 (that merely lost me on fluff levels, but still kept me buying books for completeness) I'm really considering not to buy SR5 and rather do what "criminals" would do: Wait till the stuff is availible on black market and get it from there just to be able to read and hate it even more.

1) Underlying, in that context, relates to the means of which data is translated into information. I.E, the logical construct that changes signal into information. It has nothing to do with your understanding of a network transmission model. Because network transmission is meaningless without the context of information.

Not coincidentally, this is why these devices are frequently called "Information Systems".

2) Transmission protocols are, like physical properties of transmission, only part of what makes up the entirety of system architecture.

Let's sidestep the Matrix and Shadowrun for a moment, and discuss implementation of a system. I will give you a scenario, and you may highlight the important components that must be considered for me to gauge the responsiveness of my system.

My system operates as follows: First a large number of queries are generated against a database. Those queries are executed. The results are formatted in a specific way. And then then results are returned to me. When calculating the responsiveness of this system, which of the following are important for me to consider:

A) Communication Medium
B) Existing Database complexity and Implementation
C) Means of Query Implementation and Complexity
D) All of the above

The answer, of course, is D. Obviously. And that's a short list of a number of other considerations. But the answer doesn't change, it's always: 'D - All of the above'. Even when a number of the implementation details match.

How it operates is a feature of implementation. Not solely physics. (Sidestepping that, of course, the filling up, shifting, and reorganization of registers is physics. So in that respect, it rides on physics. But not all logical implementations of physical properties are equal.)

Why would someone implement something as "simple" as point to point communication any way that couldn't be adequately handled locally in less time than it could be handled by distribution? There is no fundamental requirement that it operates that way except that someone decided to make it operate that way. Period. Full stop.

Are you used to using systems where wired, direct connections are "faster"? Sure. Why? Because the system architecture was implemented that way based on the needs and requirements of that system.

In 2075? That needs and requirements of that system appear to have changed.



An old, salty engineer I used to work with would frequently say: Should is a big word.

So, yes. As we understand most directly connected systems today, authentication is not something we're especially concerned about. And Authentication was definitely something we were not especially worried about when we began building distributed systems to begin with.

To that, all I can say is that how we believe things should work is often corrected in tomorrow's white paper. Regardless, it is an example of where overhead to seemingly "simple" things can be found.

Edit: Actually, I will add that building a system with a single interface (physical or logic) is frequently easier than building and maintaining more than one interface. So, in a world where the Matrix is "safe" and only criminals or ludites go "offline", you may be able to make a business case for interface design consolidation.

-Wired_SR_AEGIS

~sigh~ This is seriously getting frustrating ...



Sorry, but last time I checked "the locical construct that changes signal into information" a.k.a. was still "above" the signal. You are not using correct vocabulary. That's part of what makes your attempts so "painful".



Not coincidentally the term "Information System" is not related to "underlying" or "transmission layers".



Unfortunately for you I nowhere restricted myself to "transmission protocols".



If you insist on yet another fruitless attempt you can certainly do so.



And now let's come back to the point where the sole difference between the compared systems was A) and you "violently" agreed that wired would be better than wi-fi wheras I even accepted that - under idealized conditions - they could be equal there as well. So now, where will the two systems in question have any difference in "responsiveness"?



Wow ... now I wonder why I even mentioned a deliberate structuring of the protocol and then asked the question how and why nobody is capable of bypassing that decision?



Which is a "stupid" assumption to begin with. It's not plausible even from within the SR universe because you have people there that are bent on bypassing protocol limitations on a much larger scale. For a last time I will ask you this: How can deckers/hackers, that are capable of hacking Matrix protocols, ad hoc cracking of encryption and building specialized hardware for that very purpose at the same be incapable of bypassing that particular design decision? The answer to that: Developer's fiat in order to maintain "combat hacking". Full stop.



Yes ... and criminals and even benovelent hackers both in our real world as well as within the SR universe are constantly trying to make that "big word" work ... and they succeed.



~grr~ Again: My assumption was and is, that all parts of the protocol - thus including authetication - would be part of the communication regardless of communication medium.



So now you're trying to tell me that the end points are not capable of successfull authentication of incomming transmissions and thus have to rely on distributed computational power to get there?
Now I wonder which protocol they use for offloading part of the authentication process and how they authenticate the Matrix' response to such requests ... No, please refrain from trying to provide an answer to that.

But since this whole communication isn't leading anywhere, I'll simply say we'll have to agree to disagree and then ignore any further comment by you on this issue, because you're not worth the hassle of increasing the headache I already got from this.