It is not magical that a device knows the other devices it is connecting to, it is a BASIC principal of routing.

Do you understand that spoof is pretending to be another device?

Yea. Spoofing is your comlink saying that it is the one with that IP address or that MAC, or in SR4 the "access ID", and it's commands should be followed. The computer version of copying some one's voice.

The thing is, since the 70's, you could slave devices using telnet, across the world and into space. Slaving isn't new, and it works from the Earth to Mars. Once you make your connection and authenticate, you are in charge of the machine. You need to be in communication range, obviously, but it doesn't matter how many times you go through routers or bounce off satellites to achieve that communication, it still works fine. Just your response slows down. Cause it takes a while for your commands to get to Mars, and it takes a while to get feedback on how well those commands worked.

I'm fine with agreeing to disagree.
Of course, none of this matters for the conversation at hand, as you could just slave all of the other devices to one of the devices you are controlling, and command with that one through the satellite uplink.

Which would be silly of course, as you can just daisy chain slaved connections as far as you want.

edit> As in, I slave all drones to the comlink in my Fed-Boeing Courier. Then I slave that link to my living persona. Launch Courier, launch rotor drones, begin playing Ritt der Walküren, and attack.

This really bugs me. Its more than just the IP, more than the MAC, it is also the kerberos ticket and the SSL key pair and everything from identification to authorization when spoofing a command works.

However, if you block ALL connections except the master, that includes any possible dynamic routers. I hate that you conveniently omit that part of the quote.

I once had a contractor who would throw technobabble at me in an attempt to build credibility. The misapplication of things like equating telneting (or even cascading telnet connections) as slaving reminds me of those days.

Doesn't the SSL key pair actually make the connection? That's what I mean about using tunneling. The devices are virtually connected through SSL, which is much more important than any kind of actual physical link. The secure socket IS the connection.

I think to put this into modern terms, it is better to refer to all 7 layers of the OSI model. You can't get that high in the connection stack if you are blocked at a lower level of the connection stack.

I counter that you are confusing part of the connection as the entity of the connection.

OK. At least we understand each other. Please correct me if I'm wrong.
My position is "if there is a VPN set up, which appears to be what "subscribing" is, you can slave devices. And you can subscribe across any number of intermediary connections."
Your position is "there must be a physical layer connection between two devices for slaving".

I'll go with mine, but can work around it OK with yours.

But only if those routers show up in the connection and "routing" does not simply retransmit the "frames" without changing anything. I've heard good arguments and quotes for both interpretations, so the takeaway message is the same as for several other things in Unwired (Swap Echo, anyone?): Discuss it with your GM beforehand

Personally, I think directly connected slaves are what RAW means, but in play the "slave around the planet" interpretation is nevertheless preferable since it means you don't have to keep track of signal ranges...

One final comment since we are now at an understanding. The rules explicitly state that routing requires a connection. A VPN requires a routed connection. You can house-rule that a device is slaveable across a VPN, but that is not RAW.

EDIT:
Layer 2 is neither physical nor route-able.
Layer 3 is filterable.
Isn't SSL at layer 5?

Broadcast is significantly different that routing, but... its your table.

I thought it was on the Network layer, but wiki says it's Presentation.

Are you just trying to bring RL technical details into the Matrix?

Just trying to get common understanding. There are several of us at my table employed in technology fields. Given this is a game, without a common understanding what is, and is not, possible becomes a source of contention at the table. Using current facts about how technology works allows us not only to envision these fanciful enhancements.

The technological people at my table actually mostly like sr4 matrix. At its core, it has a solid enough basis in real world to be understandable and predictable and enough fanciful stuff to make it fun (and challenging) to be a hacker.

I like the matrix, but a more textbook approach would be nice. You can't throw in a term like "uplink" without defining it and expect everyone to think it means the same thing.
I think the only way to accomplish this is to have ALL of the matrix info in the matrix book. Don't be clever with the "kids, this is how the matrix works" magazine type articles, just lay everything out in order. Reprint the stuff from the BBB in with it. Don't reference back and forth. Check out GURPS Magic as an example. There is a magic section in the main GURPS book, and there is a magic book. If you have the magic book, you never need to look at the magic section of the main book again.

I agree with this sentiment 100%...

Problem is, at times the Matrix also takes significant breaks from reality...and when it does, the books don't carry announcements like "warning, the procedure for intercepting decrypted traffic is TOTALLY at odds with RL tech". We don't know when the analogy to real technology works and when it waves a giant red herring in our faces, so the best solution is to consider it a red herring all the time.

Now of course, this reasoning is also colored by an individual background, so...