Hi
Hello there, again ;-)
1. Hacking devices within hosts
Does that "interaction" imply that a hacker has automatically have access to that device, if he successfully hacked a host?
It mean that before you are allowed to take the Control Device action or the Change Icon action etc you need to gain User access on the network the device is connected to and before you are allowed to take the Reboot Device action or the Format Device action etc you need to gain Admin access on the network the device is connected to.
You only need to gain User or Admin access on the network the device is connected to,
once.
You
don't take another test in order to gain User or Admin access on individual devices within the network a second time. Once you have User or Admin access you are no longer considered 'outside'. Once you have some level of access on this network you also have the same level of access on all devices connected to this network.
2.) Hacking devices within a PAN:
Does "first" mean, that a hacker has to do a "second" hack (bruteforce/backdoor) to grant access to a slaved device within an PAN?
It means that in order to take actions that require User or Admin access on a device connected to a network you first need to gain User or Admin access, once, on the network.
It does
not mean that you need to take a second
second Brute Force / Probe+Backdoor Entry test on each individual device.
In addition, the German version of the core rulebook says something interesting about device rating that is not part of the original rules. Maybe it will be added later one?
I was bit surprised when you mentioned this the first time over at reddit. This is how it used to work in the 5th edition, but this got deliberately changed for the 6E (both firewall and data processing values of commlinks have been explicitly lowered compared to 5th edition to reflect that devices normally have a firewall or processing rating of 0 in this edition).
The most likely scenario here is that it seem as if the translator (German Pegasus) took some liberty when translating and, wrongly, reintroduced this rule to 6E while doing the German translation.
This will have the undesired side effects of making commlink defense would become weaker than almost all your other devices and as a result there would be no longer any point of slaving devices to your commlink at all. This goes against Rule as Intended by the author of the English version as well as Rules as Written in the the English version.
Basically if I don't see an icon of the gun I could do a matrix perception looking for the hidden icon for the gun so I can hack it, but not that the GM now has to roll an opposed test for every hidden icon in their immediate area.
In previous edition you had to take a matrix perception test to spot individual devices and if the individual device was running silent then the test was opposed but if not and within 100 meters then spotting was automatic. After that you also had to mark individual devices before you could control them.
In this edition matrix perception is always opposed, but in this edition all icons are also normally immediately obvious and does not require a test to be spotted at all (there is no 100 meter ´'limit' in this edition).
Also, in this edition you either run your entire network silent or you don't. If a network is trying to hide (silent running) then you need to take one opposed matrix perception test to spot the entire silent running network.
In this edition you no longer mark individual devices, in this edition you gain User or Admin access on the entire network (including all individual devices that are part of the network) just once.