[penllawen]

I suppose we could resolve it like this- normal users can connect instantaneously, but hacking in 2080 relies on sensitive covert timing channels.

I have an answer for this. Edit to add - in my own, homebrew, non-canonical fluff.

I split the Matrix into two: first, a "local mesh" of peer-to-peer mesh-networked nodes, with a range of 1/2km or so but with noticeable signal degradation across that range. Then an "upper grid" or "backbone" of superfast fibre/other sci-fi gunk. The two are linked by "uplink nodes" (aka "beanstalks".)

Traffic hops from device to device around the local mesh, mostly constrained to very short range (only barely better than line of sight.) It's high bandwidth, high frequency wireless. If the traffic is headed to somewhere outside of local mesh range, it goes to an uplink node, where it is swept up onto the backbone. On the backbone, bandwidth is functionally infinite, and you can communicate with anything.

The uplink nodes are high-rating hosts that are very well patrolled by GOD. The backbone is routinely and exhaustively traffic-snooped by GOD and its pet semi-sapient AIs. It's very, very difficult to sneak any hacking traffic on there (in game terms, you get huge Overwatch Score every second.)

But the local mesh is the wild west; a huge, messy soup of devices running different protocols and software versions and each with their own forest of vulnerabilities. This is where hackers can do their thing, but only when they are within local mesh range of their target. And they are disrupted by eg. wireless blocking paint.

Hosts are in two types too. Cloud hosts are attached to the backbone, and are therefore close to impregnable; not only do they have their own defences, but the backbone itself protects them. But the latency is too great to form WANs with cloud hosts. So there are also local hosts, which operate entirely in the context of the local mesh. These are your things that run security for a facility, the R&D terminals, local secure file storage, etc. These are shadowrunner's targets.

In addition, the corps do not 100% trust GOD. They worry that some demiGOD from a different corp will use the traffic inspection to steal their secrets. So there's an entire underground world of hidden local hosts, data couriers with chips in their head, anonymous people with briefcases full of chips, dark fibre links between sites -- all to shuttle around secrets away from the prying eyes of GOD. And all of them vulnerable to our beloved shadowrunners.

I understand that resolving this with real-word internet is futile, but it is certainly fun

I think so too. That's why I've been writing my attempt at the fluff!

[Xenon]

I have an answer for this.

(which one perhaps should point out is the full matrix rewrite you mentioned earlier, rather than how we think it actually is supposed to work in 6th edition)

[penllawen]

I have an answer for this.

(which one perhaps should point out is the full matrix rewrite you mentioned earlier, rather than how we think it actually is supposed to work in 6th edition)

Good call - will edit to call that out more explicitly.

[Lethrendis]

I assumed that according to the CRB, some of the hosts are attached to physical hardware at a specific location (p. 185). So I thought that when they cover a specific location, they have a physical boundary and cannot be infinite.

So do these physical hosts have their range limited by noise similar to PANs or are they as unlimited as purely virtual hosts?

[penllawen]

And you don't want to slave everything to your biggest Host because it means you got a single point of failure, plus with that many users policing it becomes a nightmare. Can you imagine Google using a single office building for their entire 100k staff?

This is a poor analogy because physical infrastructure and software infrastructure don't scale the same way. You can easily build distributed systems that function as a single front end but are not a single point of failure as they’re are made up of components with failover. Google has many of these. This is the closest real world equivalent to an SR host, I believe.

[Stainless Steel Devil Rat]

I assumed that according to the CRB, some of the hosts are attached to physical hardware at a specific location (p. 185). So I thought that when they cover a specific location, they have a physical boundary and cannot be infinite.

So do these physical hosts have their range limited by noise similar to PANs or are they as unlimited as purely virtual hosts?

I believe that it's for the best to treat hosts the same whether they have a physical location or not.  That way it doesn't matter if they've got a physical location... and if so, where (in the entire world) that physical location might actually be.

[markelphoenix]

I assumed that according to the CRB, some of the hosts are attached to physical hardware at a specific location (p. 185). So I thought that when they cover a specific location, they have a physical boundary and cannot be infinite.

So do these physical hosts have their range limited by noise similar to PANs or are they as unlimited as purely virtual hosts?

I believe that it's for the best to treat hosts the same whether they have a physical location or not.  That way it doesn't matter if they've got a physical location... and if so, where (in the entire world) that physical location might actually be.

So, back to my Mars question. Ares operatives could be connected and using the Ares host for protection, instead of having to rely upon a local PAN protected by local Matrix support?

My main thing is, if there is no distance to a host, then no corp team would ever need to worry about PAN security, given that all their gear would be secured by their corp host, no?

[Stainless Steel Devil Rat]

I assumed that according to the CRB, some of the hosts are attached to physical hardware at a specific location (p. 185). So I thought that when they cover a specific location, they have a physical boundary and cannot be infinite.

So do these physical hosts have their range limited by noise similar to PANs or are they as unlimited as purely virtual hosts?

I believe that it's for the best to treat hosts the same whether they have a physical location or not.  That way it doesn't matter if they've got a physical location... and if so, where (in the entire world) that physical location might actually be.

So, back to my Mars question. Ares operatives could be connected and using the Ares host for protection, instead of having to rely upon a local PAN protected by local Matrix support?

My main thing is, if there is no distance to a host, then no corp team would ever need to worry about PAN security, given that all their gear would be secured by their corp host, no?

With regards to your main thing:

yeah.  basically.  the optimal thing would usually be to just slave everything on-site to the host.  However, NPCs aren't immune from doing sub-optimal things.  There are any number of reasons why the non-optimal choice is made.

With regards to the Mars example:

The rules don't really represent matrix dead zones very well.  The middle of Siberia, the bottom of the Ocean floor, the dark side of the Moon... they all may as well be Mars as far as the Matrix rules are concerned.  If a scene takes place where Matrix connectivity is not a given, you're going to have to address it with extra and/or house rules.  Maybe the research facility on the bottom of the ocean has a hardline running up to a buoy so all that salt water noise is bypassed.  Maybe if you're in Siberia the matrix flat out doesn't work, unless you have a sat dish to contact the nearest matrix signal (up in orbit) and since all your comms must route through it, you're still at -5 dice due to distance to hack someone 1 meter away.  Or not.  Basically the point is all of these situations are beyond the scope of the matrix basic assumption, and would need to be addressed on a case by case basis.  And likely a writer by writer, and/or GM by GM basis...

[Xenon]

So, back to my Mars question...

While noise due to distance is the most common type of noise:

The most common source of noise is distance from your target

...noise is also caused by being at remote locations:

...but there are other causes... Static zones are ... far away from civilization (the middle of a desert, the north pole, adrift in the Pacific, etc.).

Spam zone or static zone:       Rating

If noise is greater than the device rating of the device you use to access the matrix with then you cannot access the matrix (including virtual hosts that you normally have zero distance to):

If noise is greater than the device rating, the device cannot access the Matrix or provide wireless bonuses.

Having said that, I think it is also fair to rule that connections to virtual hosts (that you normally have zero distance to while you are on Earth) will still suffer noise due to distance if you are not on Earth.

[penllawen]

While noise due to distance is the most common type of noise ... If noise is greater than the device rating of the device you use to access the matrix with then you cannot access the matrix (including virtual hosts that you normally have zero distance to):

Well, now we're back to the interpretation that says a rating 2 commlink in Seattle cannot call someone in Tacoma, because noise from 1-10 km is 3.

[Xenon]

Well, now we're back to the interpretation that says a rating 2 commlink in Seattle cannot call someone in Tacoma, because noise from 1-10 km is 3.

In 5th edition this was clearly not an issue.
Because of two reasons:

• Noise due to distance in the 5th edition acted as a negative dice pool modifier. And there was no test associated with Send Message to begin with
• While high levels of noise caused the device to lose its wireless bonus functionality (but depending on your reading it could also be read as if it lost its connection to the matrix as a whole) - noise due to distance did explicitly not contribute to this

Having said that, depending on your reading, in 6th edition this might or might not be an issue.

• Unlike 5th edition, in 6th edition it is clear that high levels of noise both mess with your connection and the device's wireless bonuses, not just your device's wireless bonus functionality
• Unlike 5th edition, noise due to distance is not explicitly excluded from this anymore

That connection is lost if noise due to distance is too great seem to be intended (for things such as remote controlling slaved drones).

But I don't know what the intent is when it comes to sending messages or making phone calls between two devices that are on different continents to be honest... perhaps you make long distance calls through global telecom hosts that have access points and relay stations at multiple places throughout the city and also covering all continents?

Perhaps Banshee could shred some light on this one.

[penllawen]

In 5th edition this was clearly not an issue.

Sure, I was referring to your SR6 quote:

If noise is greater than the device rating of the device you use to access the matrix with then you cannot access the matrix (including virtual hosts that you normally have zero distance to):

If noise is greater than the device rating, the device cannot access the Matrix or provide wireless bonuses.

That doesn't say 'noise due to distance'; and it does say 'cannot access the Matrix'. Your quote seems clear-cut to me that when noise (from any source) is higher than the device rating, the device rating cannot do any Matrix actions, regardless of whether there is a test involved or not. If you can't access the Matrix, I don't see how you can do Matrix things.

[Xenon]

If you can't access the Matrix, I don't see how you can do Matrix things.

Agreed.
(Which would mean you typically can not connect to a host on Earth if you are located on Mars, which was the question I was trying to answer).

But as I said, maybe the intent for long distance calls is to utilize the fact that hosts can have access points and that there does not seem to be any 'distance' within a host. That way you perhaps only need to have a signal that is strong enough to reach the nearest access point (that is perhaps just a few hundred meters away) rather than a signal that is strong enough to reach the target commlink on that other continent.

Or perhaps it is intended to work in some other way. Perhaps you have an idea of your own?

I am not the author here and the book doesn't say (but I find it plausible that you can make long distance phone calls even 50+ years from now).

[penllawen]

Perhaps you have an idea of your own?

Nothing beyond "toss big chunks of this out and start over", I am afraid.

(but I find it plausible that you can make long distance phone calls even 50+ years from now).

Not having functioning long distance phone calls is not the kind of dystopia I want, either.

[Banshee]

I have stated this before elsewhere, but specifically on things like normal comm calls and other everyday issues that come up, it's a matter of being routed through your local service provider.

Making a call to Tokyo from Seattle for most people is a matter of connecting to the the local hub (so most likely zero noise or at least minimal) then that hub (a virtual host most likely) relays the call to it's Tokyo counterpart that then connects to your "target" on the other end.

Now obviously shadowrunnner hackers don't want this potential monitoring that comes with such service so hacking over that distance comes with a severe price... thus we have the noise rules as written.